Arnoldo Frigessi

• Liu, Qinghua; Vitelli, Valeria; Mannino, Carlo; Frigessi, Arnoldo & Scheel, Ida (2026). Pseudo-Mallows for Efficient Probabilistic Preference Learning. Proceedings of Machine Learning Research (PMLR). 306, p. 78648–78680. doi: https:/proceedings.mlr.press/v306/liu26ee.html. Full text in Research Archive
• Azimzade, Youness; Haugen, Mads; Kristensen, Vessela N.; Frigessi, Arnoldo & Luque, Alvaro Kohn (2026). Explainable machine learning-guided integrated multiomics analysis reveals macrophage-driven immune suppression in breast cancer. Nature Communications. 17(1). doi: 10.1038/s41467-026-73617-9. Full text in Research Archive Show summary

  Despite thorough characterizations of cellular compositions within the breast tumor microenvironment (TME), their implications for disease progression and patient prognosis remain poorly understood. Unraveling these effects is vital for identifying potential targets to improve treatment outcomes. In this study, we devise an explainable machine learning (XML) pipeline to scrutinize the associations between TME cellular constituents and relapse-free survival (RFS). By applying this pipeline to estimated cell fractions in the METABRIC and TCGA datasets and comparing these results with associations to pathological complete response (pCR) after neoadjuvant chemotherapy (NAC), we create a comprehensive catalog of the TME’s role based on 5000 patient samples. Our findings reveal an unexpected dichotomy in which macrophages correlate positively with pCR but negatively with RFS, particularly within estrogen receptor-positive (ER+) and Luminal A and B (LumA/B) cancer subtypes. We show that this pattern is driven by heterogeneity in breast tumors characterized by increasing levels of macrophage infiltration. Through imaging mass cytometry (IMC) data analysis, we find that macrophages tend to accumulate in the vicinity of HLA-ABChi epithelial cells as their frequency increases in tumor tissues and that they also express elevated levels of HLA-ABC protein. In both IMC and single-cell RNA sequencing (scRNA-seq) data, we uncover a significant association between these HLA-ABChi macrophages and regulatory and exhausted T cells (TReg and TEx), suggesting their involvement in immune suppression, likely by creating a chronically activated immunosuppressive TME. Subsequent cell-cell communication analysis predicts interactions between HLA-ABChi macrophages and TEx cells via the ligands SIGLEC9, ALCAM, and CSF1, and with TReg cells through APP, ANGPTL4, and SIGLEC9 signaling. Considering the clinical relevance of macrophages in ER+ (LumA/B) subtypes, this work enhances the characterization of macrophage-associated immune suppression in these tumors and identifies potential targets for immunomodulatory strategies.

• Günther, Felix; Brustad, Hilde Kjelgaard; Frigessi, Arnoldo & Britton, Tom (2026). Quantifying the impact of social activities on SARS-CoV-2 transmission using Google mobility reports. PLOS Global Public Health. 6(5 May). doi: 10.1371/journal.pgph.0006368. Full text in Research Archive
• Moreira-Soares, Maurício; Fossen, Erlend I. F.; Bilbao-Jayo, Aritz; Almeida, Aitor; Lopez-Perez, Laura & Alonso, Itziar [Show all 16 contributors for this article] (2026). Joint probability framework for the development and validation of a prognostic model for the conditional outcome of quality of life: a retrospective study in historical European cohorts of survivors of head and neck cancer. The Lancet Digital Health. 8(5). doi: 10.1016/j.landig.2025.100976. Full text in Research Archive
• Kaiser, Daniel; Frigessi, Arnoldo; Ramezani-Kebrya, Ali & Ricaud, Benjamin (2026). CogniLoad: A Synthetic Natural Language Reasoning Benchmark With Tunable Length, Intrinsic Difficulty, and Distractor Density. International Conference on Learning Representations (ICLR). doi: https:/openreview.net/forum?id=0Sex2H5Jnn. Full text in Research Archive Show summary

  Current benchmarks for long-context reasoning in Large Language Models (LLMs) often blur critical factors like intrinsic task complexity, distractor interference, and task length. To enable more precise failure analysis, we introduce CogniLoad, a novel synthetic benchmark grounded in Cognitive Load Theory (CLT). CogniLoad generates natural-language logic puzzles with independently tunable parameters that reflect CLT's core dimensions: intrinsic difficulty ($d$) controls intrinsic load; distractor-to-signal ratio ($\rho$) regulates extraneous load; and task length ($N$) serves as an operational proxy for conditions demanding germane load. Evaluating 22 SotA reasoning LLMs, CogniLoad reveals distinct performance sensitivities, identifying task length as a dominant constraint and uncovering varied tolerances to intrinsic complexity and U-shaped responses to distractor ratios. By offering systematic, factorial control over these cognitive load dimensions, CogniLoad provides a reproducible, scalable, and diagnostically rich tool for dissecting LLM reasoning limitations and guiding future model development.

• Kaiser, Daniel; Ramezani-Kebrya, Ali; Frigessi, Arnoldo & Ricaud, Benjamin (2025). Decomposing Reasoning Efficiency in Large Language Models. International Conference on Learning Representations (ICLR). doi: https:/openreview.net/forum?id=mym1D6mIZR. Full text in Research Archive Show summary

  While large language models (LLMs) are typically evaluated on accuracy metrics alone, real-world deployment requires careful control of computational efficiency. Building on the CogniLoad benchmark, we introduce a unified efficiency metric (i.e., correct answers per 1'000 output tokens) enabling direct cross‑model comparison. We further provide an interpretable factorization into context robustness, logic robustness, and token appetite. Evaluating 15 state‑of‑the‑art reasoning LLMs on CogniLoad, we find that some models fail due to logic errors, others consume excessive tokens, and a few hit context limits. Tokens are an imperfect but practical proxy for computation load, permitting consistent comparisons across closed and open models. By decomposing overall efficiency into actionable components, our framework identifies concrete targets for improving LLM reasoning efficiency.

• Smajdor, Anna; Frigessi, Arnoldo & Zucknick, Manuela (2025). WHAT IS RIGHT FOR ‘THIS’ PATIENT? AUTONOMY, TRUST AND TRANSPARENCY IN THE ERA OF MEDICAL AI. Síntesis: Revista de Filosofia. ISSN 2452-4476. 52(164), p. 611–611. doi: 10.20911/21769389v52n164p611/2025. Full text in Research Archive Show summary

  In this paper we compare the capabilities and limitations of AI algorithms and human doctors in the context of treatment recommendations. Ethical concerns have been raised in the literature about transparency, accountability, and interpretability in relation to AI algorithms in the medical context. Patients’ autonomy is not likely to be respected if treatment recommendations are opaque or incomprehensible. Medical AI algorithms are programmed to prioritise predefined goals that may not align with patients’ preferences, potentially undermining their autonomy without their awareness. Yet human doctors often face similar challenges, such as reliance on clinical trial data with endpoints that may not be fully understood by patients. We conclude that there are opportunities now to integrate patient autonomy considerations into AI algorithms at the design stage, and that these goals and any relevant weightings are made explicit to end users, including both doctors and patients. Keywords: Artificial intelligence. Autonomy Alignment. Medical ethics.

• Lengle, Julie Marie; Zhang, Chi; Frigessi, Arnoldo & Andersen, Lene Frost (2025). Nutritionally adequate, healthy, and climate-friendly diets following the Nordic Nutrition Recommendations 2023: an optimization study for Norway. Frontiers in Nutrition. 12, p. 1–16. doi: 10.3389/fnut.2025.1485287. Full text in Research Archive Show summary

  Background: Reducing red meat consumption is an effective tactic for decreasing environmental impact of diets while maintaining nutritional adequacy, healthiness, and overall consumer acceptability. Still, dietary change in favor of plant foods is a controversial climate mitigation measure, especially in the Nordic region where agri-food heritage is linked to ruminant husbandry. Objective: In this study we aimed to explore sustainable diets for the Norwegian context by (1) investigating the environmental impacts of nutritionally optimized diets following the Nordic Nutrition Recommendations 2023 (NNR2023), (2) estimating potential for environmental impact reduction across scenarios of meat and legume consumption, and (3) identifying nutritional challenges. Methods: Quadratic optimization was employed to minimize departure from the average observed Norwegian diet while meeting nutrient, health, and carbon footprint constraints. The diet of Norwegian adults was estimated based on results from the national dietary survey Norkost 3. Global warming potential (GWP), freshwater and marine eutrophication, terrestrial acidification, water use, and transformation and use of land were calculated using data from the Norwegian Life Cycle Assessment Food Database version 01. Diets were optimized to meet NNR2023 nutrition and health recommendations for nutrients and food groups. Optimizations were first run without constraints on GWP, for three diet scenarios: (1) nutrients and health-based targets for food amounts ( NNR2023 ), (2) nutrients and health-based targets for food amounts with ruminant meat ≥ observed intake (62 g/day) ( Ruminant ), and (3) nutrients and health-based targets for food amounts with legumes content ≥40 g/day (Legumes). Then, GWP constraints were applied in 5% increments until no solution was found. The optimal diet for each scenario was defined as the diet with the largest feasible reduction in GWP (NNR2023+/Ruminant+/Legumes+). Results: Optimizing the diet to meet nutrient and health constraints alone resulted in a modest decrease in GWP (NNR2023); retaining ruminant meat consumption (Ruminant) impeded the reduction (−9% vs. 0%). Diets following NNR2023 nutrient and health constraints alone were feasible up until a 30% reduction in GWP (NNR2023+). A 35% reduction in GWP was achieved when legumes were added to the diet (Legumes+), while diets retaining 62 g of ruminant meat were not identified beyond a 15% reduction in GWP (Ruminant+). Sodium and selenium were the strongest limiting constraints in all scenarios. Diets with a 40% reduction in GWP were identified when nutrient constraints were lowered from the Recommended Intake to the Average Requirement (NNR2023+/Legumes+). Reductions in GWP coincided with reductions in all measured environmental indicators except marine eutrophication. Conclusion: The NNR2023 guidelines outline diets that have generally lower environmental impacts than the average Norwegian diet, though outcomes depend on distribution of meat and legume consumption in the diet. Regardless of degree of environmental impact reduction, diets following NNR2023 guidelines will require significant dietary changes compared to observed intake, including an increase in consumption of fruits, vegetables, and grains, and a strong decrease in consumption of red meat, total meat, and discretionary foods. Preventing the model from removing any ruminant meat from the diet limited GWP reduction to 15% and induced considerable changes in intake of other food groups, especially a decrease in other types of meat.

• Pagui, Euloge Clovis Kenne; Brant, Simon Boge; Svendsen, Kristian; Frigessi, Arnoldo; Helland, Åslaug & Thoresen, Steinar Østerbø [Show all 7 contributors for this article] (2025). Comparative Effectiveness of Atezolizumab, Nivolumab and Pembrolizumab in Second‐Line Treatment of Advanced Non–Small Cell Lung Cancer. Pharmacoepidemiology and Drug Safety. ISSN 1053-8569. 34(7). doi: 10.1002/pds.70181. Full text in Research Archive Show summary

  ABSTRACT Purpose This observational study aimed to assess the comparative effectiveness of immunotherapy (IO) in the second‐line setting of advanced non–small cell lung cancer (aNSCLC). Methods Population‐based data was obtained from the Cancer Registry of Norway (CRN) and linked with the hospital data from the Norwegian Patient Registry (NPR) and treatment data from the Norwegian Prescribed Drug Registry (NorPD). Adult patients with aNSCLC diagnosis in Norway in 2015–2021 were included. The final cohort included patients using IO in the second‐line setting after first‐line treatment with platinum‐containing doublet chemotherapy. We performed a target trial emulating key eligibility criteria commonly used in aNSCLC randomised clinical trials (RCTs) and constructed three pair‐wise 1:1 propensity score (PS)‐matched comparison cohorts: atezolizumab versus pembrolizumab, atezolizumab versus nivolumab and nivolumab versus pembrolizumab. Age, sex, morphology, ECOG performance status, PD‐L1 expression level, chemoradiation use and comorbidities were included in the PS. The primary outcome was overall survival (OS). Patients were followed until experiencing death, emigration or end of study. A Cox proportional hazards regression model was used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). Results Analyses included a total of 447 patients, 206 treated with atezolizumab, 71 nivolumab and 170 pembrolizumab. The median follow‐up time was 8.4 (IQR: 3.4–20.7) months. We found no evidence of significant differences in OS in any of the three pair‐wise comparisons; atezolizumab versus pembrolizumab (HR: 0.86; 95% CI: 0.65–1.12), atezolizumab versus nivolumab (HR: 0.97; 95% CI: 0.68–1.37) and nivolumab versus pembrolizumab (HR: 0.99; 95% CI: 0.67–1.49). Conclusion The study suggests no evidence of significant differences between the three IO drugs atezolizumab, nivolumab and pembrolizumab when used in second‐line aNSCLC. However, wide CIs and moderate sample sizes limit definitive conclusions. These results highlight the need for larger studies to confirm potential equivalence or detect meaningful differences.

• Griesbauer, Elisabeth Maria; Czado, Claudia; Frigessi, Arnoldo & Haff, Ingrid Hobæk (2025). TVineSynth: A Truncated C-Vine Copula Generator of Synthetic Tabular Data to Balance Privacy and Utility. Proceedings of Machine Learning Research (PMLR). 258, p. 3511–3519. doi: https:/raw.githubusercontent.com/mlresearch/v258/main/assets/griesbauer25a/griesbauer25a.pdf. Full text in Research Archive Show summary

  We propose TVineSynth, a vine copula based synthetic tabular data generator, which is designed to balance privacy and utility, using the vine tree structure and its truncation to do the trade-off. Contrary to synthetic data generators that achieve DP by globally adding noise, TVineSynth performs a controlled approximation of the estimated data generating distribution, so that it does not suffer from poor utility of the resulting synthetic data for downstream prediction tasks. TVineSynth introduces a targeted bias into the vine copula model that, combined with the specific tree structure of the vine, causes the model to zero out privacy-leaking dependencies while relying on those that are beneficial for utility. Privacy is here measured with membership (MIA) and attribute inference attacks (AIA). Further, we theoretically justify how the construction of TVineSynth ensures AIA privacy under a natural privacy measure for continuous sensitive attributes. When compared to competitor models, with and without DP, on simulated and on real-world data, TVineSynth achieves a superior privacy-utility balance.

• Pérez, Tomás Fernando Varnet; Øvergaard, Kristin Romvig; Frigessi, Arnoldo & Biele, Guido (2025). Long-term effect of pharmacological treatment on academic achievement of Norwegian children diagnosed with ADHD: a target trial emulation. International Journal of Epidemiology. ISSN 0300-5771. 54(2), p. 1–11. doi: 10.1093/ije/dyaf010. Full text in Research Archive Show summary

  Background Attention-deficit/hyperactivity disorder (ADHD) is one of the most commonly diagnosed mental disorders in children. For many patients, treatment involves long-term medication in order to reduce symptoms, regulate behaviour, and, hopefully, improve school performance and achievement. However, there is little to no evidence to support a long-term effect on the latter complex outcomes. Methods We utilize a target trial framework to emulate a pretest–posttest control group design and estimate the intention-to-treat effect of ADHD medication on national test scores in children diagnosed with ADHD born between 2000 and 2007 in Norway. The data were obtained through linkage of Norwegian registries (NorPD, Norwegian Prescription Database; NPR, Norwegian Patient Registry; KUHR, Database for Control and Payment of Health Reimbursement; SSB, Statistics Norway; MBRN, Medical Birth Registry of Norway). Results The resulting analytic sample size consisted of 8548 children diagnosed with ADHD, with about 9% missingness in their grade eight national test scores. We find that initiating ADHD medication had a slight positive average effect on national test scores for all three domains: English, numeracy, and reading [standardized mean differences: 0.037 (95% compatibility interval (CI95), −0.003; 0.076), 0.063 (CI95, 0.016; 0.111), 0.071 (CI95, 0.030; 0.111), respectively]. Conclusion We conclude that the estimated long-term average effect of ADHD medication on learning, as measured by the Norwegian national tests, is not clinically relevant. Study strengths include the use of real-world data on ecologically valid and relevant outcomes and the robustness of results across model specifications. Limitations include possibility of unobserved confounding and lack of prescription data.

• Schmiester, Leonard; Brasó-Maristany, Fara; González-Farré, Blanca; Pascual, Tomás; Gavilá, Joaquín & Tekpli, Xavier [Show all 11 contributors for this article] (2024). Computational Model Predicts Patient Outcomes in Luminal B Breast Cancer Treated with Endocrine Therapy and CDK4/6 Inhibition. Clinical Cancer Research. ISSN 1078-0432. 30(17), p. 3779–3787. doi: 10.1158/1078-0432.CCR-24-0244. Full text in Research Archive Show summary

  Purpose: Development of a computational biomarker to predict, prior to treatment, the response to CDK4/6 inhibition (CDK4/6i) in combination with endocrine therapy in patients with breast cancer. Experimental Design: A mechanistic mathematical model that accounts for protein signaling and drug mechanisms of action was developed and trained on extensive, publicly available data from breast cancer cell lines. The model was built to provide a patient-specific response score based on the expression of six genes (CCND1, CCNE1, ESR1, RB1, MYC, and CDKN1A). The model was validated in five independent cohorts of 148 patients in total with early-stage or advanced breast cancer treated with endocrine therapy and CDK4/6i. Response was measured either by evaluating Ki67 levels and PAM50 risk of relapse (ROR) after neoadjuvant treatment or by evaluating progression-free survival (PFS). Results: The model showed significant association with patient’s outcomes in all five cohorts. The model predicted high Ki67 [area under the curve; AUC (95% confidence interval, CI) of 0.80 (0.64–0.92), 0.81 (0.60–1.00) and 0.80 (0.65–0.93)] and high PAM50 ROR [AUC of 0.78 (0.64–0.89)]. This observation was not obtained in patients treated with chemotherapy. In the other cohorts, patient stratification based on the model prediction was significantly associated with PFS [hazard ratio (HR) = 2.92 (95% CI, 1.08–7.86), P = 0.034 and HR = 2.16 (1.02 4.55), P = 0.043]. Conclusions: A mathematical modeling approach accurately predicts patient outcome following CDK4/6i plus endocrine therapy that marks a step toward more personalized treatments in patients with Luminal B breast cancer.

• Chan, Yat Hin; Rø, Gunnar; Midtbø, Jørgen E.; Ruscio, Francesco Di; Watle, Sara Sofie Viksmoen & Juvet, Lene Kristine [Show all 19 contributors for this article] (2024). Modeling geographic vaccination strategies for COVID-19 in Norway. PLoS Computational Biology. ISSN 1553-734X. 20(1). doi: 10.1371/journal.pcbi.1011426. Full text in Research Archive Show summary

  Vaccination was a key intervention in controlling the COVID-19 pandemic globally. In early 2021, Norway faced significant regional variations in COVID-19 incidence and prevalence, with large differences in population density, necessitating efficient vaccine allocation to reduce infections and severe outcomes. This study explored alternative vaccination strategies to minimize health outcomes (infections, hospitalizations, ICU admissions, deaths) by varying regions prioritized, extra doses prioritized, and implementation start time. Using two models (individual-based and meta-population), we simulated COVID-19 transmission during the primary vaccination period in Norway, covering the first 7 months of 2021. We investigated alternative strategies to allocate more vaccine doses to regions with a higher force of infection. We also examined the robustness of our results and highlighted potential structural differences between the two models. Our findings suggest that early vaccine prioritization could reduce COVID-19 related health outcomes by 8% to 20% compared to a baseline strategy without geographic prioritization. For minimizing infections, hospitalizations, or ICU admissions, the best strategy was to initially allocate all available vaccine doses to fewer high-risk municipalities, comprising approximately one-fourth of the population. For minimizing deaths, a moderate level of geographic prioritization, with approximately one-third of the population receiving doubled doses, gave the best outcomes by balancing the trade-off between vaccinating younger people in high-risk areas and older people in low-risk areas. The actual strategy implemented in Norway was a two-step moderate level aimed at maintaining the balance and ensuring ethical considerations and public trust. However, it did not offer significant advantages over the baseline strategy without geographic prioritization. Earlier implementation of geographic prioritization could have more effectively addressed the main wave of infections, substantially reducing the national burden of the pandemic.

• Wu, Chenyu; Gunnarsson, Einar Bjarki; Myklebust, Even Moa; Luque, Alvaro Kohn; Tadele, Dagim Shiferaw & Enserink, Jorrit [Show all 9 contributors for this article] (2024). Using birth-death processes to infer tumor subpopulation structure from live-cell imaging drug screening data. PLoS Computational Biology. ISSN 1553-734X. 20(3). doi: 10.1371/journal.pcbi.1011888. Full text in Research Archive Show summary

  Tumor heterogeneity is a complex and widely recognized trait that poses significant challenges in developing effective cancer therapies. In particular, many tumors harbor a variety of subpopulations with distinct therapeutic response characteristics. Characterizing this heterogeneity by determining the subpopulation structure within a tumor enables more precise and successful treatment strategies. In our prior work, we developed PhenoPop, a computational framework for unravelling the drug-response subpopulation structure within a tumor from bulk high-throughput drug screening data. However, the deterministic nature of the underlying models driving PhenoPop restricts the model fit and the information it can extract from the data. As an advancement, we propose a stochastic model based on the linear birth-death process to address this limitation. Our model can formulate a dynamic variance along the horizon of the experiment so that the model uses more information from the data to provide a more robust estimation. In addition, the newly proposed model can be readily adapted to situations where the experimental data exhibits a positive time correlation. We test our model on simulated data (in silico) and experimental data (in vitro), which supports our argument about its advantages.

• Yildiz, Tugba Akman; Arendt, Lisa M.; Geisler, Jürgen; Kristensen, Vessela N.; Frigessi, Arnoldo & Luque, Alvaro Kohn (2024). Modeling of Mouse Experiments Suggests that Optimal Anti-Hormonal Treatment for Breast Cancer is Diet-Dependent. Bulletin of Mathematical Biology. ISSN 0092-8240. 86:42(4), p. 1–33. doi: 10.1007/s11538-023-01253-1. Full text in Research Archive Show summary

  Estrogen receptor positive breast cancer is frequently treated with anti-hormonal treatment such as aromatase inhibitors (AI). Interestingly, a high body mass index has been shown to have a negative impact on AI efficacy, most likely due to disturbances in steroid metabolism and adipokine production. Here, we propose a mathematical model based on a system of ordinary differential equations to investigate the effect of high-fat diet on tumor growth. We inform the model with data from mouse experiments, where the animals are fed with high-fat or control (normal) diet. By incorporating AI treatment with drug resistance into the model and by solving optimal control problems we found differential responses for control and high-fat diet. To the best of our knowledge, this is the first attempt to model optimal anti-hormonal treatment for breast cancer in the presence of drug resistance. Our results underline the importance of considering high-fat diet and obesity as factors influencing clinical outcomes during anti-hormonal therapies in breast cancer patients.

• Moreira-Soares, Maurício; Fossen, Erlend Ignacio Fleck; Taylor, Katherine; Singer, Susanne; Hurley, Katrina & Thomas, Steve [Show all 16 contributors for this article] (2024). Quality of life in head and neck cancer survivors: the Big Data for Quality of Life study. medRxiv. doi: 10.1101/2024.02.02.24302118. Full text in Research Archive
• Fossen, Erlend Ignacio Fleck; Moreira-Soares, Maurício; Leblanc, Marissa Erin; Frigessi, Arnoldo; Hovig, Eivind & Lopez-Perez, Laura [Show all 20 contributors for this article] (2024). External validation of prognostic and predictive gene signatures in head and neck cancer patients. medRxiv. doi: 10.1101/2024.09.24.24314278. Full text in Research Archive
• Moreira-Soares, Maurício; Fossen, Erlend Ignacio Fleck; Bilbao-Jayo, Aritz; Almeida, Aitor; Lopez-Perez, Laura & Alonso, Itziar [Show all 17 contributors for this article] (2024). Joint probability approach for prognostic prediction of conditional outcomes: application to quality of life in head and neck cancer survivors. medRxiv. Full text in Research Archive
• Kamineni, Meghana; Engø-Monsen, Kenth; Midtbø, Jørgen E.; Forland, Frode; Blasio, Birgitte Freiesleben De & Frigessi, Arnoldo [Show all 7 contributors for this article] (2023). Effects of non-compulsory and mandatory COVID-19 interventions on travel distance and time away from home, Norway, 2021. Eurosurveillance. ISSN 1025-496X. 28(17). doi: 10.2807/1560-7917.ES.2023.28.17.2200382. Full text in Research Archive Show summary

  Background Given the societal, economic and health costs of COVID-19 non-pharmaceutical interventions (NPI), it is important to assess their effects. Human mobility serves as a surrogate measure for human contacts and compliance with NPI. In Nordic countries, NPI have mostly been advised and sometimes made mandatory. It is unclear if making NPI mandatory further reduced mobility. Aim We investigated the effect of non-compulsory and follow-up mandatory measures in major cities and rural regions on human mobility in Norway. We identified NPI categories that most affected mobility. Methods We used mobile phone mobility data from the largest Norwegian operator. We analysed non-compulsory and mandatory measures with before–after and synthetic difference-in-differences approaches. By regression, we investigated the impact of different NPI on mobility. Results Nationally and in less populated regions, time travelled, but not distance, decreased after follow-up mandatory measures. In urban areas, however, distance decreased after follow-up mandates, and the reduction exceeded the decrease after initial non-compulsory measures. Stricter metre rules, gyms reopening, and restaurants and shops reopening were significantly associated with changes in mobility. Conclusion Overall, distance travelled from home decreased after non-compulsory measures, and in urban areas, distance further decreased after follow-up mandates. Time travelled reduced more after mandates than after non-compulsory measures for all regions and interventions. Stricter distancing and reopening of gyms, restaurants and shops were associated with changes in mobility.

• Storvik, Geir Olve; Palomares, Alfonso Diz-Lois; Engebretsen, Solveig; Rø, Gunnar Øyvind Isaksson; Engø-Monsen, Kenth & Kristoffersen, Anja Bråthen [Show all 8 contributors for this article] (2023). A sequential Monte Carlo approach to estimate a time-varying reproduction number in infectious disease models: the Covid-19 case. Journal of the Royal Statistical Society, Series A (Statistics in Society). ISSN 0964-1998. 186(4), p. 616–632. doi: 10.1093/jrsssa/qnad043. Full text in Research Archive Show summary

  Abstract The Covid-19 pandemic has required most countries to implement complex sequences of non-pharmaceutical interventions, with the aim of controlling the transmission of the virus in the population. To be able to take rapid decisions, a detailed understanding of the current situation is necessary. Estimates of time-varying, instantaneous reproduction numbers represent a way to quantify the viral transmission in real time. They are often defined through a mathematical compartmental model of the epidemic, like a stochastic SEIR model, whose parameters must be estimated from multiple time series of epidemiological data. Because of very high dimensional parameter spaces (partly due to the stochasticity in the spread models) and incomplete and delayed data, inference is very challenging. We propose a state-space formalization of the model and a sequential Monte Carlo approach which allow to estimate a daily-varying reproduction number for the Covid-19 epidemic in Norway with sufficient precision, on the basis of daily hospitalization and positive test incidences. The method was in regular use in Norway during the pandemics and appears to be a powerful instrument for epidemic monitoring and management.

• Kabeta, Teshome; Holst, Rene; Wondafrash, Beyene; Frigessi, Arnoldo & Gebremariam, Mekdes (2023). Determinants of household dietary diversity in rural Ethiopia: A household panel study. Journal of Agriculture and Food Research. 12. doi: 10.1016/j.jafr.2023.100550. Full text in Research Archive Show summary

  Background Diet diversification ensures nutritional security and hence helps in meeting nutrient requirements in humans. Despite its importance, diet diversification remains a challenge, especially in rural communities in developing countries. This study aimed to identify the determinants of DD in rural Ethiopia. Methods In this study, we used an agricultural and nutrition household panel data. This study collected data from 1200 households twice a year for two years in nine districts. The household dietary diversity score (HDDS) was used as the outcome measure. Several potential demographic, social, economic, and geographic determinants of HDDS were assessed. Mixed effects truncated Poisson regression was used to identify which of these determinants were associated with HDDS at household level. Result The HDDS value ranged from 1 to 10 with a mean of 5.52 and standard deviation (SD) of 1.54. A unit increment in the variety of crop production gave a 3.4% [95% CI: 1.9%, 4.9%,] increase in relative change (RC) of HDDS. A birr increases in income generated from livestock products resulted in a 2.6% [1.5%, 3.7%] increase in RC of HDDS. The post-harvesting season contributed to a 6.4% [3.6%, 9.3%] increase in RC of HDDS compared to the pre-harvesting. Households headed by a woman had an HDDS that was −7.9% [-12.5%, −3.3%] lower in RC of HDDS than male-lead households. An additional year of education of the household head resulted in a 1.2% [0.7%, 1.7%] increase in RC of HDDS. As the mean women's decision-making score increased by one point, the RC in HDDS increased by 3.8% [1.1%, 6.4%]. No significant association was found between the time taken to reach local markets and HDDS. Conclusion This study identified economic, educational, social, and seasonal factors associated with HDDS, which should be considered when planning interventions aimed at improving HDDS in rural Ethiopia or elsewhere.

• Engebretsen, Solveig; Palomares, Alfonso Diz-Lois; Rø, Gunnar Øyvind Isaksson; Kristoffersen, Anja Bråthen; Lindstrøm, Jonas Christoffer & Engø-Monsen, Kenth [Show all 15 contributors for this article] (2023). A real-time regional model for COVID-19: Probabilistic situational awareness and forecasting. PLoS Computational Biology. ISSN 1553-734X. 19(1). doi: 10.1371/journal.pcbi.1010860. Full text in Research Archive Show summary

  The COVID-19 pandemic is challenging nations with devastating health and economic consequences. The spread of the disease has revealed major geographical heterogeneity because of regionally varying individual behaviour and mobility patterns, unequal meteorological conditions, diverse viral variants, and locally implemented non-pharmaceutical interventions and vaccination roll-out. To support national and regional authorities in surveilling and controlling the pandemic in real-time as it unfolds, we here develop a new regional mathematical and statistical model. The model, which has been in use in Norway during the first two years of the pandemic, is informed by real-time mobility estimates from mobile phone data and laboratory-confirmed case and hospitalisation incidence. To estimate regional and time-varying transmissibility, case detection probabilities, and missed imported cases, we developed a novel sequential Approximate Bayesian Computation method allowing inference in useful time, despite the high parametric dimension. We test our approach on Norway and find that three-week-ahead predictions are precise and well-calibrated, enabling policy-relevant situational awareness at a local scale. By comparing the reproduction numbers before and after lockdowns, we identify spatially heterogeneous patterns in their effect on the transmissibility, with a stronger effect in the most populated regions compared to the national reduction estimated to be 85% (95% CI 78%-89%). Our approach is the first regional changepoint stochastic metapopulation model capable of real time spatially refined surveillance and forecasting during emergencies.

• Luque, Alvaro Kohn; Myklebust, Even Moa; Tadele, Dagim Shiferaw; Giliberto, Mariaserena; Schmiester, Leonard & Noory, Jasmine [Show all 15 contributors for this article] (2023). Phenotypic deconvolution in heterogeneous cancer cell populations using drug-screening data. Cell Reports Methods. 3(3). doi: 10.1016/j.crmeth.2023.100417. Full text in Research Archive Show summary

  Tumor heterogeneity is an important driver of treatment failure in cancer since therapies often select for drug-tolerant or drug-resistant cellular subpopulations that drive tumor growth and recurrence. Profiling the drug-response heterogeneity of tumor samples using traditional genomic deconvolution methods has yielded limited results, due in part to the imperfect mapping between genomic variation and functional characteristics. Here, we leverage mechanistic population modeling to develop a statistical framework for profiling phenotypic heterogeneity from standard drug-screen data on bulk tumor samples. This method, called PhenoPop, reliably identifies tumor subpopulations exhibiting differential drug responses and estimates their drug sensitivities and frequencies within the bulk population. We apply PhenoPop to synthetically generated cell populations, mixed cell-line experiments, and multiple myeloma patient samples and demonstrate how it can provide individualized predictions of tumor growth under candidate therapies. This methodology can also be applied to deconvolution problems in a variety of biological settings beyond cancer drug response.

• Parviero, Riccardo; Hellton, Kristoffer Herland; Haug, Ola; Engø-Monsen, Kenth; Rognebakke, Hanne Therese Wist & Canright, Geoffrey [Show all 8 contributors for this article] (2022). An agent-based model with social interactions for scalable probabilistic prediction of performance of a new product. International Journal of Information Management Data Insights. 2(2). doi: 10.1016/j.jjimei.2022.100127. Full text in Research Archive Show summary

  Understanding the spreading process of new products provides valuable knowledge that can be used for effective marketing. The ability to make early prediction of success or failure is a great advantage in innovation processes. Extending current literature in a novel way, we propose a data-driven agent-based methodology that accomplishes this task. Inference and predictions are based on short-time observations of the product adoption history and knowledge of the social network of consumers. We model and predict adoptions at the agent level as driven by unobserved peer-to-peer influence and external factors such as marketing. The method compares interaction between consumers and general campaigns, and quantifies the importance of characteristics of customers and their social relations. Our computationally efficient method is demonstrated by analyzing real data, predicting the process far into the future using data from a short period after launch, and validated by simulation experiments on a true full-scale communication network.

• Page, Christian Magnus; Nøst, Therese Haugdahl; Djordjilović, Vera; Thoresen, Magne; Frigessi, Arnoldo & Sandanger, Torkjel M [Show all 7 contributors for this article] (2022). Pre-diagnostic DNA methylation in blood leucocytes in cutaneous melanoma; a nested case–control study within the Norwegian Women and Cancer cohort. Scientific Reports. 12(1). doi: 10.1038/s41598-022-18585-y. Full text in Research Archive Show summary

  The prognosis of cutaneous melanoma depends on early detection, and good biomarkers for melanoma risk may provide a valuable tool to detect melanoma development at a pre-clinical stage. By studying the epigenetic profile in pre-diagnostic blood samples of melanoma cases and cancer free controls, we aimed to identify DNA methylation sites conferring melanoma risk. DNA methylation was measured at 775,528 CpG sites using the Illumina EPIC array in whole blood in incident melanoma cases (n = 183) and matched cancer-free controls (n = 183) in the Norwegian Women and Cancer cohort. Phenotypic information and ultraviolet radiation exposure were obtained from questionnaires. Epigenome wide association (EWAS) was analyzed in future melanoma cases and controls with conditional logistic regression, with correction for multiple testing using the false discovery rate (FDR). We extended the analysis by including a public data set on melanoma (GSE120878), and combining these different data sets using a version of covariate modulated FDR (AdaPT). The analysis on future melanoma cases and controls did not identify any genome wide significant CpG sites (0.85 ≤ padj ≤ 0.99). In the restricted AdaPT analysis, 7 CpG sites were suggestive at the FDR level of 0.15. These CpG sites may potentially be used as pre-diagnostic biomarkers of melanoma risk.

• Vitelli, Valeria; Fleischer, Thomas; Ankill, Jørgen; Arjas, Elja; Frigessi, Arnoldo & Kristensen, Vessela N. [Show all 7 contributors for this article] (2022). Transcriptomic pan-cancer analysis using rank-based Bayesian inference. Molecular Oncology. ISSN 1574-7891. 00(0), p. 1–16. doi: 10.1002/1878-0261.13354. Full text in Research Archive Show summary

  The analysis of whole genomes of pan-cancer data sets provides a challenge for researchers, and we contribute to the literature concerning the identification of robust subgroups with clear biological interpretation. Specifically, we tackle this unsupervised problem via a novel rank-based Bayesian clustering method. The advantages of our method are the integration and quantification of all uncertainties related to both the input data and the model, the probabilistic interpretation of final results to allow straightforward assessment of the stability of clusters leading to reliable conclusions, and the transparent biological interpretation of the identified clusters since each cluster is characterized by its top-ranked genomic features. We applied our method to RNA-seq data from cancer samples from 12 tumor types from the Cancer Genome Atlas. We identified a robust clustering that mostly reflects tissue of origin but also includes pan-cancer clusters. Importantly, we identified three pan-squamous clusters composed of a mix of lung squamous cell carcinoma, head and neck squamous carcinoma, and bladder cancer, with different biological functions over-represented in the top genes that characterize the three clusters. We also found two novel subtypes of kidney cancer that show different prognosis, and we reproduced known subtypes of breast cancer. Taken together, our method allows the identification of robust and biologically meaningful clusters of pan-cancer samples.

• Eide, Simen; Leslie, David S. & Frigessi, Arnoldo (2022). Dynamic slate recommendation with gated recurrent units and Thompson sampling. Data mining and knowledge discovery. ISSN 1384-5810. 36, p. 1756–1786. doi: 10.1007/s10618-022-00849-w. Full text in Research Archive Show summary

  We consider the problem of recommending relevant content to users of an internet platform in the form of lists of items, called slates. We introduce a variational Bayesian Recurrent Neural Net recommender system that acts on time series of interactions between the internet platform and the user, and which scales to real world industrial situations. The recommender system is tested both online on real users, and on an offline dataset collected from a Norwegian web-based marketplace, FINN.no, that is made public for research. This is one of the first publicly available datasets which includes all the slates that are presented to users as well as which items (if any) in the slates were clicked on. Such a data set allows us to move beyond the common assumption that implicitly assumes that users are considering all possible items at each interaction. Instead we build our likelihood using the items that are actually in the slate, and evaluate the strengths and weaknesses of both approaches theoretically and in experiments. We also introduce a hierarchical prior for the item parameters based on group memberships. Both item parameters and user preferences are learned probabilistically. Furthermore, we combine our model with bandit strategies to ensure learning, and introduce ‘in-slate Thompson sampling’ which makes use of the slates to maximise explorative opportunities. We show experimentally that explorative recommender strategies perform on par or above their greedy counterparts. Even without making use of exploration to learn more effectively, click rates increase simply because of improved diversity in the recommended slates.

• Azulay, Naomi; Olsen, Roy Bjørkholt; Nielsen, Christopher Sivert; Stubhaug, Audun; Jenssen, Trond Geir & Schirmer, Henrik [Show all 9 contributors for this article] (2022). Reduced heart rate variability is related to the number of metabolic syndrome components and manifest diabetes in the sixth Tromsø study 2007–2008. Scientific Reports. 12(1). doi: 10.1038/s41598-022-15824-0. Full text in Research Archive Show summary

  Both diabetes mellitus (DM) and the metabolic syndrome (MetS) are associated with autonomic neuropathy, which predisposes to cardiac events and death. Measures of heart rate variability (HRV) can be used to monitor the activity of the autonomic nervous system (ANS), and there are strong indications that HRV can be used to study the progression of ANS-related diabetes complications. This study aims to investigate diferences in HRV in healthy, MetS and diabetic populations. Based on 7880 participants from the sixth health survey in Tromsø (Tromsø 6, 2007–2008), we found a signifcant negative association between the number of MetS components and HRV as estimated from short-term pulse wave signals (PRV). This decrease in PRV did not appear to be linear, instead it leveled of after the third component, with no signifcant diference in PRV between the MetS and DM populations. There was a signifcant negative association between HbA1c and PRV, showing a decrease in PRV occurring already within the normal HbA1c range. The MetS and DM populations are diferent from healthy controls with respect to PRV, indicating impaired ANS in both conditions. In the future, a study with assessment of PRV measurements in relation to prospective cardiovascular events seems justifed.

• Jalali, Neda; Brustad, Hilde Kjelgaard; Frigessi, Arnoldo; Macdonald, Emily Ann; Meijerink, Hinta & Feruglio, Siri Laura [Show all 10 contributors for this article] (2022). Increased household transmission and immune escape of the SARS-CoV-2 Omicron compared to Delta variants. Nature Communications. 13(1), p. 1–5. doi: 10.1038/s41467-022-33233-9. Full text in Research Archive Show summary

  Understanding the epidemic growth of the novel SARS-CoV-2 Omicron variant is critical for public health. We compared the ten-day secondary attack rate (SAR) of the Omicron and Delta variants in households using Norwegian contact tracing data, December 2021 - January 2022. Omicron SAR was higher than Delta, with a relative risk (RR) of 1.41 (95% CI 1.27-1.56). We observed increased susceptibility to Omicron infection in household contacts compared to Delta, independent of contacts’ vaccination status. Among three-dose vaccinated contacts, the mean SAR was lower for both variants. We found increased Omicron transmissibility from primary cases to contacts in all vaccination groups, except 1-dose vaccinated, compared to Delta. Omicron SAR of three-dose vaccinated primary cases was high, 46% vs 11 % for Delta. In conclusion, three-dose vaccinated primary cases with Omicron infection can efficiently spread in households, while three-dose vaccinated contacts have a lower risk of being infected by Delta and Omicron.

• Ravindran, Vandana; Wagoner, Jessica; Athanasiadis, Paschalis; Hartigh, Andreas B Den; Sidorova, Julia & Ianevski, Aleksandr [Show all 11 contributors for this article] (2022). Discovery of host-directed modulators of virus infection by probing the SARS-CoV-2-host protein-protein interaction network. Briefings in Bioinformatics. ISSN 1467-5463. 23(6). doi: 10.1093/bib/bbac456. Full text in Research Archive Show summary

  The ongoing coronavirus disease 2019 (COVID-19) pandemic has highlighted the need to better understand virus–host interactions. We developed a network-based method that expands the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2)–host protein interaction network and identifies host targets that modulate viral infection. To disrupt the SARS-CoV-2 interactome, we systematically probed for potent compounds that selectively target the identified host proteins with high expression in cells relevant to COVID-19. We experimentally tested seven chemical inhibitors of the identified host proteins for modulation of SARS-CoV-2 infection in human cells that express ACE2 and TMPRSS2. Inhibition of the epigenetic regulators bromodomain-containing protein 4 (BRD4) and histone deacetylase 2 (HDAC2), along with ubiquitin-specific peptidase (USP10), enhanced SARS-CoV-2 infection. Such proviral effect was observed upon treatment with compounds JQ1, vorinostat, romidepsin and spautin-1, when measured by cytopathic effect and validated by viral RNA assays, suggesting that the host proteins HDAC2, BRD4 and USP10 have antiviral functions. We observed marked differences in antiviral effects across cell lines, which may have consequences for identification of selective modulators of viral infection or potential antiviral therapeutics. While network-based approaches enable systematic identification of host targets and selective compounds that may modulate the SARS-CoV-2 interactome, further developments are warranted to increase their accuracy and cell-context specificity.

• Pesonen, Henri; Simola, Umberto; Luque, Alvaro Kohn; Vuollekoski, Henri; Lai, Xiaoran & Frigessi, Arnoldo [Show all 11 contributors for this article] (2022). ABC of the future. International Statistical Review. ISSN 0306-7734. doi: 10.1111/insr.12522. Full text in Research Archive Show summary

  Approximate Bayesian computation (ABC) has advanced in two decades from a seminal idea to a practically applicable inference tool for simulator-based statistical models, which are becoming increasingly popular in many research domains. The computational feasibility of ABC for practical applications has been recently boosted by adopting techniques from machine learning to build surrogate models for the approximate likelihood or posterior and by the introduction of a general-purpose software platform with several advanced features, including automated parallelisation. Here we demonstrate the strengths of the advances in ABC by going beyond the typical benchmark examples and considering real applications in astronomy, infectious disease epidemiology, personalised cancer therapy and financial prediction. We anticipate that the emerging success of ABC in producing actual added value and quantitative insights in the real world will continue to inspire a plethora of further applications across different fields of science, social science and technology.

• Lai, Xiaoran; Taskén, Håkon Austlid; Mo, Torgeir; Funke, Simon Wolfgang; Frigessi, Arnoldo & Rognes, Marie Elisabeth [Show all 7 contributors for this article] (2021). A scalable solver for a stochastic, hybrid cellular automaton model of personalized breast cancer therapy. International Journal for Numerical Methods in Biomedical Engineering. ISSN 2040-7939. 38(1). doi: 10.1002/cnm.3542. Full text in Research Archive Show summary

  Mathematical modeling and simulation is a promising approach to personalizedcancer medicine. Yet, the complexity, heterogeneity and multi-scale nature ofcancer pose significant computational challenges. Coupling discrete cell-basedmodels with continuous models using hybrid cellular automata (CA) is a power-ful approach for mimicking biological complexity and describing the dynamicalexchange of information across different scales. However, when clinically rele-vant cancer portions are taken into account, such models become computation-ally very expensive. While efficient parallelization techniques for continuousmodels exist, their coupling with discrete models, particularly CA, necessitatesmore elaborate solutions. Building upon FEniCS, a popular and powerful scien-tific computing platform for solving partial differential equations, we developedparallel algorithms to link stochastic CA with differential equations (https://bitbucket.org/HTasken/cansim). The algorithms minimize the communicationbetween processes that share CA neighborhood values while also allowing forreproducibility during stochastic updates. We demonstrated the potential of oursolution on a complex hybrid cellular automaton model of breast cancer treatedwith combination chemotherapy. On a single-core processor, we obtained nearlylinear scaling with an increasing problem size, whereas weak parallel scalingshowed moderate growth in solving time relative to increase in problem size.Finally, we applied the algorithm to a problem that is 500 times larger than previ-ous work, allowing us to run personalized therapy simulations based on hetero-geneous cell density and tumor perfusion conditions estimated from magneticresonance imaging data on an unprecedented scale

• Lindstrøm, Jonas Christoffer; Engebretsen, Solveig; Kristoffersen, Anja Bråthen; Rø, Gunnar Øyvind Isaksson; Palomares, Alfonso Diz-Lois & Engø-Monsen, Kenth [Show all 14 contributors for this article] (2021). Increased transmissibility of the alpha SARS-CoV-2 variant: evidence from contact tracing data in Oslo, January to February 2021. Infectious Diseases. ISSN 2374-4235. 54(1), p. 72–77. doi: 10.1080/23744235.2021.1977382. Full text in Research Archive Show summary

  Background: Information about the contagiousness of new SARS-CoV-2 variants, including the alpha lineage, and how they spread in various locations is essential. Country-specific estimates are needed because local interventions influence transmissibility. Methods: We analysed contact tracing data from Oslo municipality, reported from January through February 2021, when the alpha lineage became predominant in Norway and estimated the relative transmissibility of the alpha lineage with the use of Poisson regression. Results: Within households, we found an increase in the secondary attack rate by 60% (95% CI 20–114%) among cases infected with the alpha lineage compared to other variants; including all close contacts, the relative increase in the secondary attack rate was 24% (95% CI −6%−43%). There was a significantly higher risk of infecting household members in index cases aged 40–59 years who were infected with the alpha lineage; we found no association between transmission and household size. Overall, including all close contacts, we found that the reproduction number among cases with the alpha lineage was increased by 24% (95% CI 0%−52%), corresponding to an absolute increase of 0.19, compared to the group of index cases infected with other variants. Conclusion: Our study suggests that households are the primary locations for rapid transmission of the new lineage alpha.

• Eide, Simen; Leslie, David; Frigessi, Arnoldo; Rishaug, Joakim; Jenssen, Helge & Verrewaere, Sofie (2021). FINN. no Slates Dataset: A new Sequential Dataset Logging Interactions, all Viewed Items and Click Responses/No-Click for Recommender Systems Research. In Corona, Humberto; Larson, Martha & Willemsen, Martijn C. (Ed.), RecSys '21 : Fifteenth ACM Conference on Recommender Systems. Association for Computing Machinery (ACM). ISSN 9781450384582. doi: 10.1145/3460231.3474607. Full text in Research Archive
• Divino, Fabio; Belay, Denekew Bitew; Keilman, Nico & Frigessi, Arnoldo (2021). Spatial modelling of Lexis mortality data. Spatial Statistics. 44. doi: 10.1016/j.spasta.2021.100532. Full text in Research Archive Show summary

  We present a spatial approach to investigate mortality data referenced over a Lexis structure. We decompose the force of mortality into two interpretable components: a Markov random field, homogeneous with respect to age, period and cohort which explains the main pattern of mortality; and a secondary component of independent shocks, accounting for additional non structured patterns. Inference is based on a hierarchical Bayesian model with Markov chain Monte Carlo computations. We present an application to data from the Human Mortality Database, with respect to Italy and Sweden, two countries with very different histories in terms of demographic and epidemiological transitions. For each country the primary surface of spatially structured mortality and the secondary surface of additional mortality are estimated. The importance of each component is valuated by the estimated value of the respective precision parameter. In both Italy and Sweden, we discovered an interesting band of extra mortality in the secondary surface across the time domain, in the age interval between 60 and 90 years, with a slightly positive slope. The band represents a significant amount of extra mortality for the elderly population, which is otherwise incompatible with a structured dynamics in age, period and cohort.

• Engebretsen, Solveig; Engø-Monsen, Kenth; Aleem, Mohammad Abdul; Gurley, Emily Suzanne; Rattalma, Arnoldo Frigessi Di & Blasio, Birgitte Freiesleben de (2020). Time-aggregated mobile phone mobility data are sufficient for modelling influenza spread: the case of Bangladesh. Journal of the Royal Society Interface. ISSN 1742-5689. 17(167). doi: 10.1098/rsif.2019.0809. Full text in Research Archive Show summary

  Human mobility plays a major role in the spatial dissemination of infectious diseases. We develop a spatio-temporal stochastic model for influenza-like disease spread based on estimates of human mobility. The model is informed by mobile phone mobility data collected in Bangladesh. We compare predictions of models informed by daily mobility data (reference) with that of models informed by time-averaged mobility data, and mobility model approximations. We find that the gravity model overestimates the spatial synchrony, while the radiation model underestimates the spatial synchrony. Using time-averaged mobility resulted in spatial spreading patterns comparable to the daily mobility model. We fit the model to 2014–2017 influenza data from sentinel hospitals in Bangladesh, using a sequential version of approximate Bayesian computation. We find a good agreement between our estimated model and the case data. We estimate transmissibility and regional spread of influenza in Bangladesh, which are useful for policy planning. Time-averaged mobility appears to be a good proxy for human mobility when modelling infectious diseases. This motivates a more general use of the time-averaged mobility, with important implications for future studies and outbreak control. Moreover, time-averaged mobility is subject to less privacy concerns than daily mobility, containing less temporal information on individual movements

• Pladsen, Arne Valebjørg; Nilsen, Gro; Rueda, Oscar M.; Aure, Miriam Ragle; Borgan, Ørnulf & Liestøl, Knut [Show all 18 contributors for this article] (2020). DNA copy number motifs are strong and independent predictors of survival in breast cancer. Communications Biology. 3. doi: 10.1038/s42003-020-0884-6. Full text in Research Archive Show summary

  Somatic copy number alterations are a frequent sign of genome instability in cancer. A precise characterization of the genome architecture would reveal underlying instability mechanisms and provide an instrument for outcome prediction and treatment guidance. Here we show that the local spatial behavior of copy number profiles conveys important information about this architecture. Six filters were defined to characterize regional traits in copy number profiles, and the resulting Copy Aberration Regional Mapping Analysis (CARMA) algorithm was applied to tumors in four breast cancer cohorts (n = 2919). The derived motifs represent a layer of information that complements established molecular classifications of breast cancer. A score reflecting presence or absence of motifs provided a highly significant independent prognostic predictor. Results were consistent between cohorts. The nonsite-specific occurrence of the detected patterns suggests that CARMA captures underlying replication and repair defects and could have a future potential in treatment stratification.

• Page, Christian; Djordjilovic, Vera; Nøst, Therese Haugdahl; Ghiasvand, Reza; Sandanger, Torkjel M & Rattalma, Arnoldo Frigessi Di [Show all 8 contributors for this article] (2020). Lifetime ultraviolet radiation exposure and DNA methylation in blood leukocytes: The Norwegian Women and Cancer Study. Scientific Reports. 10, p. 1–8. doi: 10.1038/s41598-020-61430-3. Full text in Research Archive Show summary

  Ultraviolet radiation (UVR) exposure is a leading cause of skin cancers and an ubiquitous environmental exposure. However, the molecular mechanisms relating UVR exposure to melanoma is not fully understood. We aimed to investigate if lifetime UVR exposure could be robustly associated to DNA methylation (DNAm). We assessed DNAm in whole blood in three data sets (n = 183, 191, and 125) from the Norwegian Woman and Cancer cohort, using Illumina platforms. We studied genome-wide DNAm, targeted analyses of CpG sites indicated in the literature, global methylation, and accelerated aging. Lifetime history of UVR exposure (residential ambient UVR, sunburns, sunbathing vacations and indoor tanning) was collected by questionnaires. We used one data set for discovery and the other two for replication. One CpG site showed a genome-wide significant association to cumulative UVR exposure (cg01884057) (pnominal = 3.96e-08), but was not replicated in any of the two replication sets (pnominal ≥ 0.42). Two CpG sites (cg05860019, cg00033666) showed suggestive associations with the other UVR exposures. We performed extensive analyses of the association between long-term UVR exposure and DNAm. There was no indication of a robust effect of past UVR exposure on DNAm.

• Bergholtz, Helga; Lien, Tonje Gulbrandsen; Swanson, David; Rattalma, Arnoldo Frigessi Di; Daidone, Maria Grazia & Tost, Jörg [Show all 8 contributors for this article] (2020). Contrasting DCIS and invasive breast cancer by subtype suggests basal-like DCIS as distinct lesions. NPJ breast cancer. 6. doi: 10.1038/s41523-020-0167-x. Full text in Research Archive Show summary

  Ductal carcinoma in situ (DCIS) is a non-invasive type of breast cancer with highly variable potential of becoming invasive and affecting mortality. Currently, many patients with DCIS are overtreated due to the lack of specific biomarkers that distinguish low risk lesions from those with a higher risk of progression. In this study, we analyzed 57 pure DCIS and 313 invasive breast cancers (IBC) from different patients. Three levels of genomic data were obtained; gene expression, DNA methylation, and DNA copy number. We performed subtype stratified analyses and identified key differences between DCIS and IBC that suggest subtype specific progression. Prominent differences were found in tumors of the basal-like subtype: Basal-like DCIS were less proliferative and showed a higher degree of differentiation than basal-like IBC. Also, core basal tumors (characterized by high correlation to the basal-like centroid) were not identified amongst DCIS as opposed to IBC. At the copy number level, basal-like DCIS exhibited fewer copy number aberrations compared with basal-like IBC. An intriguing finding through analysis of the methylome was hypermethylation of multiple protocadherin genes in basal-like IBC compared with basal-like DCIS and normal tissue, possibly caused by long range epigenetic silencing. This points to silencing of cell adhesion-related genes specifically in IBC of the basal-like subtype. Our work confirms that subtype stratification is essential when studying progression from DCIS to IBC, and we provide evidence that basal-like DCIS show less aggressive characteristics and question the assumption that basal-like DCIS is a direct precursor of basal-like invasive breast cancer.

• Crispino, Marta; Arjas, Elja; Vitelli, Valeria; Barrett, Natasha & Rattalma, Arnoldo Frigessi Di (2019). A Bayesian Mallows approach to nontransitive pair comparison data: How human are sounds? Annals of Applied Statistics. ISSN 1932-6157. 13(1), p. 492–519. doi: 10.1214/18-AOAS1203. Full text in Research Archive Show summary

  We are interested in learning how listeners perceive sounds as having human origins. An experiment was performed with a series of electronically synthesized sounds, and listeners were asked to compare them in pairs. We propose a Bayesian probabilistic method to learn individual preferences from nontransitive pairwise comparison data, as happens when one (or more) individual preferences in the data contradicts what is implied by the others. We build a Bayesian Mallows model in order to handle nontransitive data, with a latent layer of uncertainty which captures the generation of preference misreporting. We then develop a mixture extension of the Mallows model, able to learn individual preferences in a heterogeneous population. The results of our analysis of the musicology experiment are of interest to electroacoustic composers and sound designers, and to the audio industry in general, whose aim is to understand how computer generated sounds can be produced in order to sound more human.

• Tekpli, Xavier; Lien, Tonje Gulbrandsen; Røssevold, Andreas Hagen; Nebdal, Daniel J.H.; Borgen, Elin & Ohnstad, Hege Oma [Show all 19 contributors for this article] (2019). An independent poor-prognosis subtype of breast cancer defined by a distinct tumor immune microenvironment. Nature Communications. 10:5499, p. 1–14. doi: 10.1038/s41467-019-13329-5. Full text in Research Archive Show summary

  How mixtures of immune cells associate with cancer cell phenotype and affect pathogenesis is still unclear. In 15 breast cancer gene expression datasets, we invariably identify three clusters of patients with gradual levels of immune infiltration. The intermediate immune infiltration cluster (Cluster B) is associated with a worse prognosis independently of known clinicopathological features. Furthermore, immune clusters are associated with response to neoadjuvant chemotherapy. In silico dissection of the immune contexture of the clusters identified Cluster A as immune cold, Cluster C as immune hot while Cluster B has a pro-tumorigenic immune infiltration. Through phenotypical analysis, we find epithelial mesenchymal transition and proliferation associated with the immune clusters and mutually exclusive in breast cancers. Here, we describe immune clusters which improve the prognostic accuracy of immune contexture in breast cancer. Our discovery of a novel independent prognostic factor in breast cancer highlights a correlation between tumor phenotype and immune contexture.

• Lai, Xiaoran; Geier, Oliver; Fleischer, Thomas; Garred, Øystein; Borgen, Elin & Funke, Simon Wolfgang [Show all 14 contributors for this article] (2019). Towards personalized computer simulation of breast cancer treatment: a multi-scale pharmacokinetic and pharmacodynamic model informed by multi-type patient data. Cancer Research. ISSN 0008-5472. 79(16), p. 4293–4304. doi: 10.1158/0008-5472.CAN-18-1804. Full text in Research Archive Show summary

  The usefulness of mechanistic models to disentangle complex multiscale cancer processes, such as treatment response, has been widely acknowledged. However, a major barrier for multiscale models to predict treatment outcomes in individual patients lies in their initialization and parametrization, which needs to reflect individual cancer characteristics accurately. In this study, we use multitype measurements acquired routinely on a single breast tumor, including histopathology, MRI, and molecular profiling, to personalize parts of a complex multiscale model of breast cancer treated with chemotherapeutic and antiangiogenic agents. The model accounts for drug pharmacokinetics and pharmacodynamics. We developed an open-source computer program that simulates cross-sections of tumors under 12-week therapy regimens and used it to individually reproduce and elucidate treatment outcomes of 4 patients. Two of the tumors did not respond to therapy, and model simulations were used to suggest alternative regimens with improved outcomes dependent on the tumor's individual characteristics. It was determined that more frequent and lower doses of chemotherapy reduce tumor burden in a low proliferative tumor while lower doses of antiangiogenic agents improve drug penetration in a poorly perfused tumor. Furthermore, using this model, we were able to correctly predict the outcome in another patient after 12 weeks of treatment. In summary, our model bridges multitype clinical data to shed light on individual treatment outcomes.

• Liu, Qinghua; Reiner, Andrew Henry; Rattalma, Arnoldo Frigessi Di & Scheel, Ida (2019). Diverse personalized recommendations with uncertainty from implicit preference data with the Bayesian Mallows model. Knowledge-Based Systems. ISSN 0950-7051. p. 1–12. doi: 10.1016/j.knosys.2019.104960. Full text in Research Archive Show summary

  Clicking data, which exists in abundance and contains objective user preference information, is widely used to produce personalized recommendations in web-based applications. Current popular recommendation algorithms, typically based on matrix factorizations, often focus on achieving high accuracy. While achieving good clickthrough rates, diversity of the recommended items is often overlooked. Moreover, most algorithms do not produce interpretable uncertainty quantifications of the recommendations. In this work, we propose the Bayesian Mallows for Clicking Data (BMCD) method, which simultaneously considers accuracy and diversity. BMCD augments clicking data into compatible full ranking vectors by enforcing all the clicked items clicked by a user to be top-ranked regardless of their rarity. User preferences are learned using a Mallows ranking model. Bayesian inference leads to interpretable uncertainties of each individual recommendation, and we also propose a method to make personalized recommendations based on such uncertainties. With a simulation study and a real life data example, we demonstrate that compared to state-of-the-art matrix factorization, BMCD makes personalized recommendations with similar accuracy, while achieving much higher level of diversity, and producing interpretable and actionable uncertainty estimation.

• Swanson, David; Lien, Tonje Gulbrandsen; Bergholtz, Helga; Sørlie, Therese & Rattalma, Arnoldo Frigessi Di (2019). A Bayesian two-way latent structure model for genomic data integration reveals few pan-genomic cluster subtypes in a breast cancer cohort. Bioinformatics. ISSN 1367-4803. 35(23), p. 4886–4897. doi: 10.1093/bioinformatics/btz381. Full text in Research Archive Show summary

  Motivation Unsupervised clustering is important in disease subtyping, among having other genomic applications. As genomic data has become more multifaceted, how to cluster across data sources for more precise subtyping is an ever more important area of research. Many of the methods proposed so far, including iCluster and Cluster of Cluster Assignments (COCAs), make an unreasonable assumption of a common clustering across all data sources, and those that do not are fewer and tend to be computationally intensive. Results We propose a Bayesian parametric model for integrative, unsupervised clustering across data sources. In our two-way latent structure model, samples are clustered in relation to each specific data source, distinguishing it from methods like COCAs and iCluster, but cluster labels have across-dataset meaning, allowing cluster information to be shared between data sources. A common scaling across data sources is not required, and inference is obtained by a Gibbs Sampler, which we improve with a warm start strategy and modified density functions to robustify and speed convergence. Posterior interpretation allows for inference on common clusterings occurring among subsets of data sources. An interesting statistical formulation of the model results in sampling from closed-form posteriors despite incorporation of a complex latent structure. We fit the model with Gaussian and more general densities, which influences the degree of across-dataset cluster label sharing. Uniquely among integrative clustering models, our formulation makes no nestedness assumptions of samples across data sources so that a sample missing data from one genomic source can be clustered according to its existing data sources. We apply our model to a Norwegian breast cancer cohort of ductal carcinoma in situ and invasive tumors, comprised of somatic copy-number alteration, methylation and expression datasets. We find enrichment in the Her2 subtype and ductal carcinoma among those observations exhibiting greater cluster correspondence across expression and CNA data. In general, there are few pan-genomic clusterings, suggesting that models assuming a common clustering across genomic data sources might yield misleading results. Availability and implementation The model is implemented in an R package called twl (‘two-way latent’), available on CRAN. Data for analysis are available within the R package.

• Liu, Qinghua; Crispino, Marta; Scheel, Ida; Vitelli, Valeria & Rattalma, Arnoldo Frigessi Di (2019). Model-based learning from preference data. Annual Review of Statistics and Its Application. ISSN 2326-8298. 6, p. 329–354. doi: 10.1146/annurev-statistics-031017-100213. Full text in Research Archive
• Engebretsen, Solveig; Engø-Monsen, Kenth; Rattalma, Arnoldo Frigessi Di & Blasio, Birgitte Freiesleben De (2019). A theoretical single-parameter model for urbanisation to study infectious disease spread and interventions. PLoS Computational Biology. ISSN 1553-734X. 15(3). doi: 10.1371/journal.pcbi.1006879. Full text in Research Archive Show summary

  The world is continuously urbanising, resulting in clusters of densely populated urban areas and more sparsely populated rural areas. We propose a method for generating spatial fields with controllable levels of clustering of the population. We build a synthetic country, and use this method to generate versions of the country with different clustering levels. Combined with a metapopulation model for infectious disease spread, this allows us to in silico explore how urbanisation affects infectious disease spread. In a baseline scenario with no interventions, the underlying population clustering seems to have little effect on the final size and timing of the epidemic. Under within-country restrictions on non-commuting travel, the final size decreases with increased population clustering. The effect of travel restrictions on reducing the final size is larger with higher clustering. The reduction is larger in the more rural areas. Within-country travel restrictions delay the epidemic, and the delay is largest for lower clustering levels. We implemented three different vaccination strategies—uniform vaccination (in space), preferentially vaccinating urban locations and preferentially vaccinating rural locations. The urban and uniform vaccination strategies were most effective in reducing the final size, while the rural vaccination strategy was clearly inferior. Visual inspection of some European countries shows that many countries already have high population clustering. In the future, they will likely become even more clustered. Hence, according to our model, within-country travel restrictions are likely to be less and less effective in delaying epidemics, while they will be more effective in decreasing final sizes. In addition, to minimise final sizes, it is important not to neglect urban locations when distributing vaccines. To our knowledge, this is the first study to systematically investigate the effect of urbanisation on infectious disease spread and in particular, to examine effectiveness of prevention measures as a function of urbanisation.

• Vitelli, Valeria; Sørensen, Øystein; Crispino, Marta; Rattalma, Arnoldo Frigessi Di & Arjas, Elja (2018). Probabilistic preference learning with the Mallows rank model. Journal of machine learning research. ISSN 1532-4435. 18, p. 1–49. Full text in Research Archive Show summary

  Ranking and comparing items is crucial for collecting information about preferences in many areas, from marketing to politics. The Mallows rank model is among the most successful approaches to analyse rank data, but its computational complexity has limited its use to a particular form based on Kendall distance. We develop new computationally tractable methods for Bayesian inference in Mallows models that work with any right-invariant distance. Our method performs inference on the consensus ranking of the items, also when based on partial rankings, such as top-k items or pairwise comparisons. We prove that items that none of the assessors has ranked do not influence the maximum a posteriori consensus ranking, and can therefore be ignored. When assessors are many or heterogeneous, we propose a mixture model for clustering them in homogeneous subgroups, with cluster-specific consensus rankings. We develop approximate stochastic algorithms that allow a fully probabilistic analysis, leading to coherent quantifications of uncertainties. We make probabilistic predictions on the class membership of assessors based on their ranking of just some items, and predict missing individual preferences, as needed in recommendation systems. We test our approach using several experimental and benchmark datasets.

• Rohrbeck, Christian; Costain, Deborah A & Rattalma, Arnoldo Frigessi Di (2018). Bayesian spatial monotonic multiple regression. Biometrika. ISSN 0006-3444. 105(3), p. 691–707. doi: 10.1093/biomet/asy019. Full text in Research Archive
• LeBlanc, Marissa; Zuber, Verena; Thompson, Wesley Kurt; Andreassen, Ole Andreas; Rattalma, Arnoldo Frigessi Di & Kulle, Bettina (2018). A correction for sample overlap in genome-wide association studies in a polygenic pleiotropy-informed framework. BMC Genomics. 19:494, p. 1–15. doi: 10.1186/s12864-018-4859-7. Full text in Research Archive Show summary

  Background There is considerable evidence that many complex traits have a partially shared genetic basis, termed pleiotropy. It is therefore useful to consider integrating genome-wide association study (GWAS) data across several traits, usually at the summary statistic level. A major practical challenge arises when these GWAS have overlapping subjects. This is particularly an issue when estimating pleiotropy using methods that condition the significance of one trait on the signficance of a second, such as the covariate-modulated false discovery rate (cmfdr). Results We propose a method for correcting for sample overlap at the summary statistic level. We quantify the expected amount of spurious correlation between the summary statistics from two GWAS due to sample overlap, and use this estimated correlation in a simple linear correction that adjusts the joint distribution of test statistics from the two GWAS. The correction is appropriate for GWAS with case-control or quantitative outcomes. Our simulations and data example show that without correcting for sample overlap, the cmfdr is not properly controlled, leading to an excessive number of false discoveries and an excessive false discovery proportion. Our correction for sample overlap is effective in that it restores proper control of the false discovery rate, at very little loss in power. Conclusions With our proposed correction, it is possible to integrate GWAS summary statistics with overlapping samples in a statistical framework that is dependent on the joint distribution of the two GWAS.

• Erango, Markos Abiso; Rattalma, Arnoldo Frigessi Di & Rosseland, Leiv Arne (2018). A three minutes supine position test reveals higher risk of spinal anesthesia induced hypotension during cesarean delivery. An observational study. [version 1; peer review: 3 approved]. F1000 Research. 7(1028), p. 1–20. doi: 10.12688/F1000RESEARCH.15142.1. Full text in Research Archive
• Rohrbeck, Christian; Eastoe, Emma F.; Rattalma, Arnoldo Frigessi Di & Tawn, Jonathan A. (2018). Extreme value modelling of water-related insurance claims. Annals of Applied Statistics. ISSN 1932-6157. 12(1), p. 246–282. doi: 10.1214/17-AOAS1081. Full text in Research Archive Show summary

  This paper considers the dependence between weather events, for example, rainfall or snow-melt, and the number of water-related property insurance claims. Weather events which cause severe damages are of general interest; decision makers want to take efficient actions against them while the insurance companies want to set adequate premiums. The modelling is challenging since the underlying dynamics vary across geographical regions due to differences in topology, construction designs and climate. We develop new methodology to improve the existing models which fail to model high numbers of claims. The statistical framework is based on both mixture and extremal mixture modelling, with the latter being based on a discretized generalized Pareto distribution. Furthermore, we propose a temporal clustering algorithm and derive new covariates which lead to a better understanding of the association between claims and weather events. The modelling of the claims, conditional on the locally observed weather events, both fit the marginal distributions well and capture the spatial dependence between locations. Our methodology is applied to three cities across Norway to demonstrate its benefits.

• Sørensen, Øystein; Hellton, Kristoffer Herland; Rattalma, Arnoldo Frigessi Di & Thoresen, Magne (2018). Covariate Selection in High-Dimensional Generalized Linear Models With Measurement Error. Journal of Computational And Graphical Statistics (JCGS). ISSN 1061-8600. 27(4), p. 739–749. doi: 10.1080/10618600.2018.1425626. Full text in Research Archive Show summary

  In many problems involving generalized linear models, the covariates are subject to measurement error. When the number of covariates p exceeds the sample size n, regularized methods like the lasso or Dantzig selector are required. Several recent papers have studied methods which correct for measurement error in the lasso or Dantzig selector for linear models in the p > n setting. We study a correction for generalized linear models, based on Rosenbaum and Tsybakov’s matrix uncertainty selector. By not requiring an estimate of the measurement error covariance matrix, this generalized matrix uncertainty selector has a great practical advantage in problems involving high-dimensional data. We further derive an alternative method based on the lasso, and develop efficient algorithms for both methods. In our simulation studies of logistic and Poisson regression with measurement error, the proposed methods outperform the standard lasso and Dantzig selector with respect to covariate selection, by reducing the number of false positives considerably. We also consider classification of patients on the basis of gene expression data with noisy measurements. Supplementary materials for this article are available online.

• Engebretsen, Solveig; Frigessi, Arnoldo; Engø-Monsen, Kenth; Furberg, Anne-Sofie; Stubhaug, Audun & Blasio, Birgitte Freiesleben De [Show all 7 contributors for this article] (2018). The peer effect on pain tolerance. Scandinavian Journal of Pain. ISSN 1877-8860. 18(3), p. 467–477. doi: 10.1515/sjpain-2018-0060. Full text in Research Archive Show summary

  Background and aims:<br> Twin studies have found that approximately half of the variance in pain tolerance can be explained by genetic factors, while shared family environment has a negligible effect. Hence, a large proportion of the variance in pain tolerance is explained by the (non-shared) unique environment. The social environment beyond the family is a potential candidate for explaining some of the variance in pain tolerance. Numerous individual traits have previously shown to be associated with friendship ties. In this study, we investigate whether pain tolerance is associated with friendship ties. <br> Methods:<br> We study the friendship effect on pain tolerance by considering data from the Tromsø Study: Fit Futures I, which contains pain tolerance measurements and social network information for adolescents attending first year of upper secondary school in the Tromsø area in Northern Norway. Pain tolerance was measured with the cold-pressor test (primary outcome), contact heat and pressure algometry. We analyse the data by using statistical methods from social network analysis. Specifically, we compute pairwise correlations in pain tolerance among friends. We also fit network autocorrelation models to the data, where the pain tolerance of an individual is explained by (among other factors) the average pain tolerance of the individual’s friends. <br> Results: <br> We find a significant and positive relationship between the pain tolerance of an individual and the pain tolerance of their friends. The estimated effect is that for every 1 s increase in friends’ average cold-pressor tolerance time, the expected cold-pressor pain tolerance of the individual increases by 0.21 s (p-value: 0.0049, sample size n=997). This estimated effect is controlled for sex. The friendship effect remains significant when controlling for potential confounders such as lifestyle factors and test sequence among the students. Further investigating the role of sex on this friendship effect, we only find a significant peer effect of male friends on males, while there is no significant effect of friends’ average pain tolerance on females in stratified analyses. Similar, but somewhat lower estimates were obtained for the other pain modalities. <br> Conclusions:<br> We find a positive and significant peer effect in pain tolerance. Hence, there is a significant tendency for students to be friends with others with similar pain tolerance. Sex-stratified analyses show that the only significant effect is the effect of male friends on males. <br> Implications:<br> Two different processes can explain the friendship effect in pain tolerance, selection and social transmission. Individuals might select friends directly due to similarity in pain tolerance, or indirectly through similarity in other confounding variables that affect pain tolerance. Alternatively, there is an influence effect among friends either directly in pain tolerance, or indirectly through other variables that affect pain tolerance. If there is indeed a social influence effect in pain tolerance, then the social environment can account for some of the unique environmental variance in pain tolerance. If so, it is possible to therapeutically affect pain tolerance through alteration of the social environment.

• Asfaw, Derbachew; Vitelli, Valeria; Sørensen, Øystein; Arjas, Elja & Frigessi, Arnoldo (2017). Time-varying rankings with the Bayesian Mallows model. Stat. 6(1), p. 14–30. doi: 10.1002/sta4.132. Full text in Research Archive Show summary

  We present new statistical methodology for analysing rank data, where the rankings are allowed to vary in time. Such data arise, for example, when the assessments are based on a performance measure of the items, which varies in time, or if the criteria, according to which the items are ranked, change in time. Items can also be absent when the assessments are made, because of delayed entry or early departure, or purely randomly. In such situations, also the dimension of the rank vectors varies in time. Rank data in a time‐dependent setting thus lead to challenging statistical problems. These problems are further complicated, from the perspective of computation, by the large dimension of the sample space consisting of all permutations of the items. Here, we focus on introducing and developing a Bayesian version of the Mallows rank model, suitable for situations in which the ranks vary in time and the assessments can be incomplete. The consequent missing data problems are handled by applying Bayesian data augmentation within Markov chain Monte Carlo. Our method is also adapted to the task of future rank prediction. The method is illustrated by analysing some aspects of a data set describing the academic performance, measured by a series of tests, of a class of high school students over a period of 4 years. © 2017 Wiley

• Fleischer, Thomas; Tekpli, Xavier; Mathelier, Anthony; Wang, Shixiong; Nebdal, Daniel J.H. & Dhakal, Hari Prasad [Show all 16 contributors for this article] (2017). DNA methylation at enhancers identifies distinct breast cancer lineages. Nature Communications. 8(1379). doi: 10.1038/s41467-017-00510-x. Full text in Research Archive Show summary

  Breast cancers exhibit genome-wide aberrant DNA methylation patterns. To investigate how these affect the transcriptome and which changes are linked to transformation or progression, we apply genome-wide expression–methylation quantitative trait loci (emQTL) analysis between DNA methylation and gene expression. On a whole genome scale, in cis and in trans, DNA methylation and gene expression have remarkably and reproducibly conserved patterns of association in three breast cancer cohorts (n = 104, n = 253 and n = 277). The expression–methylation quantitative trait loci associations form two main clusters; one relates to tumor infiltrating immune cell signatures and the other to estrogen receptor signaling. In the estrogen related cluster, using ChromHMM segmentation and transcription factor chromatin immunoprecipitation sequencing data, we identify transcriptional networks regulated in a cell lineage-specific manner by DNA methylation at enhancers. These networks are strongly dominated by ERα, FOXA1 or GATA3 and their targets were functionally validated using knockdown by small interfering RNA or GRO-seq analysis after transcriptional stimulation with estrogen.

• Aure, Miriam Ragle; Vitelli, Valeria; Jernström, Sandra Johanna; Kumar, Surendra; Krohn, Marit & Due, Eldri Undlien [Show all 35 contributors for this article] (2017). Integrative clustering reveals a novel split in the luminal A subtype of breast cancer with impact on outcome. Breast Cancer Research. ISSN 1465-5411. 19(44). doi: 10.1186/s13058-017-0812-y. Full text in Research Archive Show summary

  Background Breast cancer is a heterogeneous disease at the clinical and molecular level. In this study we integrate classifications extracted from five different molecular levels in order to identify integrated subtypes. Methods Tumor tissue from 425 patients with primary breast cancer from the Oslo2 study was cut and blended, and divided into fractions for DNA, RNA and protein isolation and metabolomics, allowing the acquisition of representative and comparable molecular data. Patients were stratified into groups based on their tumor characteristics from five different molecular levels, using various clustering methods. Finally, all previously identified and newly determined subgroups were combined in a multilevel classification using a “cluster-of-clusters” approach with consensus clustering. Results Based on DNA copy number data, tumors were categorized into three groups according to the complex arm aberration index. mRNA expression profiles divided tumors into five molecular subgroups according to PAM50 subtyping, and clustering based on microRNA expression revealed four subgroups. Reverse-phase protein array data divided tumors into five subgroups. Hierarchical clustering of tumor metabolic profiles revealed three clusters. Combining DNA copy number and mRNA expression classified tumors into seven clusters based on pathway activity levels, and tumors were classified into ten subtypes using integrative clustering. The final consensus clustering that incorporated all aforementioned subtypes revealed six major groups. Five corresponded well with the mRNA subtypes, while a sixth group resulted from a split of the luminal A subtype; these tumors belonged to distinct microRNA clusters. Gain-of-function studies using MCF-7 cells showed that microRNAs differentially expressed between the luminal A clusters were important for cancer cell survival. These microRNAs were used to validate the split in luminal A tumors in four independent breast cancer cohorts. In two cohorts the microRNAs divided tumors into subgroups with significantly different outcomes, and in another a trend was observed. Conclusions The six integrated subtypes identified confirm the heterogeneity of breast cancer and show that finer subdivisions of subtypes are evident. Increasing knowledge of the heterogeneity of the luminal A subtype may add pivotal information to guide therapeutic choices, evidently bringing us closer to improved treatment for this largest subgroup of breast cancer.

• Linder, Dennis; Frigessi, Arnoldo; Piaserico, Stefano & Keilman, Nico (2017). Simulating the life course of psoriasis patients: the interplay between therapy intervention and marital status. Journal of the European Academy of Dermatology and Venereology. ISSN 0926-9959. 32(1), p. 62–67. doi: 10.1111/jdv.14567. Full text in Research Archive Show summary

  Background Psoriasis, a chronic relapsing inflammatory disease affecting primarily the skin, shows multiple comorbidities including depression, cardiovascular diseases and other relevant conditions. Psoriasis patients experience social isolation, job loss, financial difficulties and partnership problems. Inversely, psychosocial impairments may negatively influence the disease course. Objective To explore the feasibility of a model describing the interaction of psychosocial and clinical factors over the life course of patients. Methods We considered only seven states for members of a hypothetical population: single and healthy, single and having a psoriasis flare, single and ‘cured’, coupled and healthy, coupled and having a psoriasis flare, coupled and ‘cured’, and dead. Transition probabilities between states were taken from the Norwegian Population Register for the healthy population and from epidemiological research articles. Clinical experience allowed adjustments on the assumed parameters. Results Our macromodel, which simulates the effect of therapy intervention on patients’ partnership status, yields a description of the transitions between the seven states. Treatment efficacy shows only a negligible effect on the chances of living with a partner. Conclusions Mathematical modelling of interactions between social and health variables is in principle feasible. However, complex models, comprising more variables (for instance: employment status, depression level, obesity etc.), are needed for more realistic simulations for the interactions studied. As increasing the number of variables leads to an exponential increase of the model’s state space, switching to micromodelling (representing each individual separately) may be necessary.

• Belay, Denekew Bitew; Kifle, Yehenew Getachew; Goshu, Ayele Taye; Gran, Jon Michael; Yewhalaw, Delenasaw & Duchateau, Luc [Show all 7 contributors for this article] (2017). Joint Bayesian modeling of time to malaria and mosquito abundance in Ethiopia. BMC Infectious Diseases. 17(1). doi: 10.1186/s12879-017-2496-4. Full text in Research Archive Show summary

  Background This paper studies the effect of mosquito abundance and malaria incidence in the last 3 weeks, and their interaction, on the hazard of time to malaria in a previously studied cohort of children in Ethiopia. Methods We model the mosquito abundance and time to malaria data jointly in a Bayesian framework. Results We found that the interaction of mosquito abundance and incidence plays a prominent role on malaria risk. We quantify and compare relative risks of various factors, and determine the predominant role of the interaction between incidence and mosquito abundance in describing malaria risk. Seasonal rain patterns, distance to a water source of the households, temperature and relative humidity are all significant in explaining mosquito abundance, and through this affect malaria risk. Conclusion Analyzing jointly the time to malaria data and the mosquito abundance allows a precise comparison of factors affecting the spread of malaria. The effect of the interaction between mosquito abundances and local presence of malaria parasites has an important effect on the hazard of time to malaria, beyond abundance alone. Each additional one km away from the dam gives an average reduction of malaria relative risk of 5.7%. The importance of the interaction between abundance and incidence leads to the hypothesis that preventive intervention could advantageously target the infectious population, in addition to mosquito control, which is the typical intervention today.

• Frigessi, Arnoldo; Buhlmann, Peter; Glad, Ingrid Kristine; Richardson, Sylvia & Vannucci, Marina (2016). Some themes in high-dimensional statistics. Abel Symposia. ISSN 2193-2808. 11, p. 1–13. doi: 10.1007/978-3-319-27099-9_1. Full text in Research Archive
• Tharmaratnam, K.; Sperrin, M.; Jaki, T.; Reppe, Sjur & Frigessi, Arnoldo (2016). Tilting the lasso by knowledge-based post-processing. BMC Bioinformatics. 17(344), p. 1–9. doi: 10.1186/s12859-016-1210-7. Full text in Research Archive Show summary

  Background It is useful to incorporate biological knowledge on the role of genetic determinants in predicting an outcome. It is, however, not always feasible to fully elicit this information when the number of determinants is large. We present an approach to overcome this difficulty. First, using half of the available data, a shortlist of potentially interesting determinants are generated. Second, binary indications of biological importance are elicited for this much smaller number of determinants. Third, an analysis is carried out on this shortlist using the second half of the data. Results We show through simulations that, compared with adaptive lasso, this approach leads to models containing more biologically relevant variables, while the prediction mean squared error (PMSE) is comparable or even reduced. We also apply our approach to bone mineral density data, and again final models contain more biologically relevant variables and have reduced PMSEs. Conclusion Our method leads to comparable or improved predictive performance, and models with greater face validity and interpretability with feasible incorporation of biological knowledge into predictive models.

• Solvang, Hiroko Kato; Frigessi, Arnoldo; Kaveh, Fatemeh; Riis, Margit; Luders, Torben & Bukholm, Ida Rashida Khan [Show all 8 contributors for this article] (2016). Gene expression analysis supports tumor threshold over 2.0 cm for T-category breast cancer. EURASIP Journal on Bioinformatics and Systems Biology. ISSN 1687-4145. 2016(6). doi: 10.1186/s13637-015-0034-5. Full text in Research Archive Show summary

  Tumor size, as indicated by the T-category, is known as a strong prognostic indicator for breast cancer. It is common practice to distinguish the T1 and T2 groups at a tumor size of 2.0 cm. We investigated the 2.0-cm rule from a new point of view. Here, we try to find the optimal threshold based on the differences between the gene expression profiles of the T1 and T2 groups (as defined by the threshold). We developed a numerical algorithm to measure the overall differential gene expression between patients with smaller tumors and those with larger tumors among multiple expression datasets from different studies. We confirmed the performance of the proposed algorithm by a simulation study and then applied it to three different studies conducted at two Norwegian hospitals. We found that the maximum difference in gene expression is obtained at a threshold of 2.2–2.4 cm, and we confirmed that the optimum threshold was over 2.0 cm, as indicated by a validation study using five publicly available expression datasets. Furthermore, we observed a significant differentiation between the two threshold groups in terms of time to local recurrence for the Norwegian datasets. In addition, we performed an associated network and canonical pathway analyses for the genes differentially expressed between tumors below and above the given thresholds, 2.0 and 2.4 cm, using the Norwegian datasets. The associated network function illustrated a cellular assembly of the genes for the 2.0-cm threshold: an energy production for the 2.4-cm threshold and an enrichment in lipid metabolism based on the genes in the intersection for the 2.0- and 2.4-cm thresholds.

• Solbergersen, LInn Cecilie; Ahmed, Ismail; Frigessi, Arnoldo; Glad, Ingrid Kristine & Richardson, Sylvia Therese Lamblin (2016). Preselection in Lasso-Type Analysis for Ultra-High Dimensional Genomic Exploration. Abel Symposia. ISSN 2193-2808. 11, p. 37–66. doi: 10.1007/978-3-319-27099-9_3. Full text in Research Archive
• Bolin, David; Frigessi, Arnoldo; Guttorp, Peter; Haug, Ola; Orskaug, Elisabeth & Scheel, Ida [Show all 7 contributors for this article] (2016). Calibrating regionally downscaled precipitation over Norway through quantile-based approaches. Advances in Statistical Climatology, Meteorology and Oceanography (ASCMO). ISSN 2364-3579. 2, p. 39–47. doi: 10.5194/ascmo-2-39-2016. Full text in Research Archive Show summary

  Dynamical downscaling of earth system models is intended to produce high-resolution climate information at regional to local scales. Current models, while adequate for describing temperature distributions at relatively small scales, struggle when it comes to describing precipitation distributions. In order to better match the distribution of observed precipitation over Norway, we consider approaches to statistical adjustment of the output from a regional climate model when forced with ERA-40 reanalysis boundary conditions. As a second step, we try to correct downscalings of historical climate model runs using these transformations built from downscaled ERA-40 data. Unless such calibrations are successful, it is difficult to argue that scenario-based downscaled climate projections are realistic and useful for decision makers. We study both full quantile calibrations and several different methods that correct individual quantiles separately using random field models. Results based on cross-validation show that while a full quantile calibration is not very effective in this case, one can correct individual quantiles satisfactorily if the spatial structure in the data are accounted for. Interestingly, different methods are favoured depending on whether ERA-40 data or historical climate model runs are adjusted.

• Crispino, Marta; Arjas, Elja; Vitelli, Valeria & Frigessi, Arnoldo (2016). Recommendation from intransitive pairwise comparisons. CEUR Workshop Proceedings. 1688. Full text in Research Archive
• Haff, Ingrid Hobæk; Aas, Kjersti; Frigessi, Arnoldo & Graziani, Virginia Lacal (2016). Structure learning in Bayesian Networks using regular vines. Computational Statistics & Data Analysis. ISSN 0167-9473. 101, p. 186–208. doi: 10.1016/j.csda.2016.03.003. Full text in Research Archive Show summary

  Learning the structure of a Bayesian Network from multidimensional data is an important task in many situations, as it allows understanding conditional (in)dependence relations which in turn can be used for prediction. Current methods mostly assume a multivariate normal or a discrete multinomial model. A new greedy learning algorithm for continuous non-Gaussian variables, where marginal distributions can be arbitrary, as well as the dependency structure, is proposed. It exploits the regular vine approximation of the model, which is a tree-based hierarchical construction with pair-copulae as building blocks. It is shown that the networks obtainable with our algorithm belong to a certain subclass of chordal graphs. Chordal graphs representations are often preferred, as they allow very efficient message passing and information propagation in intervention studies. It is illustrated through several examples and real data applications that the possibility of using non-Gaussian margins and a nonnon-linear dependency structure outweighs the restriction to chordal graphs.

• Sørensen, Øystein; Frigessi, Arnoldo & Thoresen, Magne (2015). Measurement error in Lasso: Impact and likelihood bias correction. Statistica sinica. ISSN 1017-0405. 25(2), p. 809–829. doi: 10.5705/ss.2013.180. Full text in Research Archive Show summary

  Regression with the lasso penalty is a popular tool for performing dimension reduction when the number of covariates is large. In many applications of the lasso, like in genomics, covariates are subject to measurement error. We study the impact of measurement error on linear regression with the lasso penalty, both analytically and in simulation experiments. A simple method of correction for measurement error in the lasso is then considered. In the large sample limit, the corrected lasso yields sign consistent covariate selection under conditions very similar to the lasso with perfect measurements, whereas the uncorrected lasso requires much more stringent conditions on the covariance structure of the data. Finally, we suggest methods to correct for measurement error in generalized linear models with the lasso penalty, which we study empirically in simulation experiments with logistic regression, and also apply to a classification problem with microarray data. We see that the corrected lasso selects less false positives than the standard lasso, at a similar level of true positives. The corrected lasso can therefore be used to obtain more conservative covariate selection in genomic analysis.

• LeBlanc, Marissa; Zuber, Verena; Kulle, Bettina; Witoelar, Aree; Zeng, Limin & Bettella, Francesco [Show all 26 contributors for this article] (2015). Identifying Novel Gene Variants in Coronary Artery Disease and Shared Genes with Several Cardiovascular Risk Factors. Circulation Research. ISSN 0009-7330. 1, p. 83–94. doi: 10.1161/CIRCRESAHA.115.306629. Full text in Research Archive
• Haff, Ingrid Hobæk; Frigessi, Arnoldo & Maraun, Douglas (2015). How well do regional climate models simulate the spatial dependence of precipitation? An application of pair-copula constructions. Journal of Geophysical Research (JGR): Atmospheres. ISSN 2169-897X. 120(7), p. 2624–2646. doi: 10.1002/2014JD022748. Full text in Research Archive Show summary

  We investigate how well a suite of regional climate models (RCMs) from the ENSEMBLES project represents the residual spatial dependence of daily precipitation. The study area we consider is a 200 km × 200 km region in south central Norway, with RCMs driven by ERA-40 boundary conditions at a horizontal resolution of approximately 25 km × 25 km. We model the residual spatial dependence with pair-copula constructions, which allows us to assess both the overall and tail dependence in precipitation, including uncertainty estimates. The selected RCMs reproduce the overall dependence rather well, though the discrepancies compared to observations are substantial. All models overestimate the overall dependence in the west-east direction. They also overestimate the upper tail dependence in the north-south direction during winter, and in the west-east direction during summer, whereas they tend to underestimate this dependence in the north-south direction in summer. Moreover, many of the climate models do not simulate the small-scale dependence patterns caused by the pronounced orography well. However, the misrepresented residual spatial dependence does not seem to affect estimates of high quantiles of extreme precipitation aggregated over a few grid boxes. The underestimation of the area-aggregated extreme precipitation is due mainly to the well-known underestimation of the univariate margins for individual grid boxes, suggesting that the correction of RCM biases in precipitation might be feasible.

• Frigessi, Arnoldo; Holden, Lars & Teigland, André (2015). (sfi)&lt;sup&gt;2&lt;/sup&gt; statistics for innovation - The experience of the Oslo centre in industrial statistics. Applied Stochastic Models in Business and Industry. ISSN 1524-1904. 31(1), p. 82–93. doi: 10.1002/asmb.2113. Full text in Research Archive
• Fairbairn, Catharine E.; Sayette, Michael; Aalen, Odd O. & Frigessi, Arnoldo (2015). Alcohol and Emotional Contagion An Examination of the Spreading of Smiles in Male and Female Drinking Groups. Clinical Psychological Science. ISSN 2167-7026. 3(5), p. 686–701. doi: 10.1177/2167702614548892. Full text in Research Archive
• Tramm, Trine; Mohammed, Hayat; Myhre, Simen; Kyndi, Marianne; Alsner, Jan & Børresen-Dale, Anne-Lise [Show all 9 contributors for this article] (2014). Development and Validation of a Gene Profile Predicting Benefit of Postmastectomy Radiotherapy in Patients with High-Risk Breast Cancer: A Study of Gene Expression in the DBCG82bc Cohort. Clinical Cancer Research. ISSN 1078-0432. 20(20), p. 5272–5280. doi: 10.1158/1078-0432.CCR-14-0458. Full text in Research Archive Show summary

  Purpose: To identify genes predicting benefit of radiotherapy in patients with high-risk breast cancer treated with systemic therapy and randomized to receive or not receive postmastectomy radiotherapy (PMRT). Experimental Design: The study was based on the Danish Breast Cancer Cooperative Group (DBCG82bc) cohort. Gene-expression analysis was performed in a training set of frozen tumor tissue from 191 patients. Genes were identified through the Lasso method with the endpoint being locoregional recurrence (LRR). A weighted gene-expression index (DBCG-RT profile) was calculated and transferred to quantitative real-time PCR (qRT-PCR) in corresponding formalin-fixed, paraffin-embedded (FFPE) samples, before validation in FFPE from 112 additional patients. Results: Seven genes were identified, and the derived DBCG-RT profile divided the 191 patients into “high LRR risk” and “low LRR risk” groups. PMRT significantly reduced risk of LRR in “high LRR risk” patients, whereas “low LRR risk” patients showed no additional reduction in LRR rate. Technical transfer of the DBCG-RT profile to FFPE/qRT-PCR was successful, and the predictive impact was successfully validated in another 112 patients. Conclusions: A DBCG-RT gene profile was identified and validated, identifying patients with very low risk of LRR and no benefit from PMRT. The profile may provide a method to individualize treatment with PMRT

• Kristensen, Vessela N.; Lingjærde, Ole Christian; Russnes, Hege Elisabeth Giercksky; Vollan, Hans Kristian Moen; Frigessi, Arnoldo & Børresen-Dale, Anne-Lise (2014). Principles and methods of integrative genomic analyses in cancer. Nature Reviews Cancer. ISSN 1474-175X. 14(5), p. 299–313. doi: 10.1038/nrc3721. Full text in Research Archive
• Fleischer, Thomas; Frigessi, Arnoldo; Johnson, Kevin C.; Edvardsen, Hege; Touleimat, Nizar & Klajic, Jovana [Show all 15 contributors for this article] (2014). Genome-wide DNA methylation profiles in progression to in situ and invasive carcinoma of the breast with impact on gene transcription and prognosis. Genome Biology. ISSN 1474-7596. 15, p. 435–435. doi: 10.1186/s13059-014-0435-x. Full text in Research Archive Show summary

  Background Ductal carcinoma in situ (DCIS) of the breast is a precursor of invasive breast carcinoma. DNA methylation alterations are thought to be an early event in progression of cancer, and may prove valuable as a tool in clinical decision making and for understanding neoplastic development. Results We generate genome-wide DNA methylation profiles of 285 breast tissue samples representing progression of cancer, and validate methylation changes between normal and DCIS in an independent dataset of 15 normal and 40 DCIS samples. We also validate a prognostic signature on 583 breast cancer samples from The Cancer Genome Atlas. Our analysis reveals that DNA methylation profiles of DCIS are radically altered compared to normal breast tissue, involving more than 5,000 genes. Changes between DCIS and invasive breast carcinoma involve around 1,000 genes. In tumors, DNA methylation is associated with gene expression of almost 3,000 genes, including both negative and positive correlations. A prognostic signature based on methylation level of 18 CpGs is associated with survival of breast cancer patients with invasive tumors, as well as with survival of patients with DCIS and mixed lesions of DCIS and invasive breast carcinoma. Conclusions This work demonstrates that changes in the epigenome occur early in the neoplastic progression, provides evidence for the possible utilization of DNA methylation-based markers of progression in the clinic, and highlights the importance of epigenetic changes in carcinogenesis.

• Scheel, Ida; Ferkingstad, Egil; Frigessi, Arnoldo; Haug, Ola; Hinnerichsen, Mikkel & Meze-Hausken, Elisabeth (2013). A Bayesian hierarchical model with spatial variable selection: the effect of weather on insurance claims. Journal of the Royal Statistical Society, Series C (Applied Statistics). ISSN 0035-9254. 62, p. 85–100. doi: 10.1111/j.1467-9876.2012.01039.x. Full text in Research Archive Show summary

  Climate change will affect the insurance industry. We develop a Bayesian hierarchical statistical approach to explain and predict insurance losses due to weather events at a local geographic scale. The number of weather-related insurance claims is modelled by combining generalized linear models with spatially smoothed variable selection. Using Gibbs sampling and reversible jump Markov chain Monte Carlo methods, this model is fitted on daily weather and insurance data from each of the 319 municipalities which constitute southern and central Norway for the period 1997–2006. Precise out-of-sample predictions validate the model. Our results show interesting regional patterns in the effect of different weather covariates. In addition to being useful for insurance pricing, our model can be used for short-term predictions based on weather forecasts and for long-term predictions based on downscaled climate models.

• Aldrin, Magne; Raastad, Ragnhild; Tvete, Ingunn Fride; Berild, Dag; Frigessi, Arnoldo & Leegaard, Truls Michael [Show all 9 contributors for this article] (2013). Antibiotic resistance in hospitals: a ward-specific random effect model in a low antibiotic consumption environment. Statistics in Medicine. ISSN 0277-6715. 32(8), p. 1407–1418. doi: 10.1002/sim.5636. Full text in Research Archive Show summary

  Association between previous antibiotic use and emergence of antibiotic resistance has been reported for several microorganisms. The relationship has been extensively studied, and although the causes of antibiotic resistance are multi-factorial, clear evidence of antibiotic use as a major risk factor exists. Most studies are carried out in countries with high consumption of antibiotics and corresponding high levels of antibiotic resistance, and currently, little is known whether and at what level the associations are detectable in a low antibiotic consumption environment. We conduct an ecological, retrospective study aimed at determining the impact of antibiotic consumption on antibiotic-resistant Pseudomonas aeruginosa in three hospitals in Norway, a country with low levels of antibiotic use. We construct a sophisticated statistical model to capture such low signals. To reduce noise, we conduct our study at hospital ward level. We propose a random effect Poisson or binomial regression model, with a reparametrisation that allows us to reduce the number of parameters. Inference is likelihood based. Through scenario simulation, we study the potential effects of reduced or increased antibiotic use. Results clearly indicate that the effects of consumption on resistance are present under conditions with relatively low use of antibiotic agents. This strengthens the recommendation on prudent use of antibiotics, even when consumption is relatively low.

• Sandve, Geir Kjetil; Gundersen, Sveinung; Johansen, Morten; Glad, Ingrid Kristine; Gunathasan, Krishanthi & Holden, Lars [Show all 21 contributors for this article] (2013). The Genomic HyperBrowser: an analysis web server for genome-scale data. Nucleic Acids Research (NAR). ISSN 0305-1048. 41(W1), p. W133–W141. doi: 10.1093/nar/gkt342. Full text in Research Archive Show summary

  The immense increase in availability of genomic scale datasets, such as those provided by the ENCODE and Roadmap Epigenomics projects, presents unprecedented opportunities for individual researchers to pose novel falsifiable biological questions. With this opportunity, however, researchers are faced with the challenge of how to best analyze and interpret their genome-scale datasets. A powerful way of representing genome-scale data is as feature-specific coordinates relative to reference genome assemblies, i.e. as genomic tracks. The Genomic HyperBrowser (http://hyperbrowser.uio.no) is an open-ended web server for the analysis of genomic track data. Through the provision of several highly customizable components for processing and statistical analysis of genomic tracks, the HyperBrowser opens for a range of genomic investigations, related to, e.g., gene regulation, disease association or epigenetic modifications of the genome.

• Løland, Anders; Huseby, Ragnar Bang; Hjort, Nils Lid & Frigessi, Arnoldo (2013). Statistical Corrections of Invalid Correlation Matrices. Scandinavian Journal of Statistics. ISSN 0303-6898. 40(4), p. 807–824. doi: 10.1111/sjos.12035. Full text in Research Archive Show summary

  Suppose estimates are available for correlations between pairs of variables but that the matrix of correlation estimates is not positive definite. In various applications, having a valid correlation matrix is important in connection with follow-up analyses that might, for example, involve sampling from a valid distribution. We present new methods for adjusting the initial estimates to form a proper, that is, nonnegative definite, correlation matrix. These are based on constructing certain pseudo-likelihood functions, formed by multiplying together exact or approximate likelihood contributions associated with the individual correlations. Such pseudo-likelihoods may then be maximized over the range of proper correlation matrices. They may also be utilized to form pseudo-posterior distributions for the unknown correlation matrix, by factoring in relevant prior information for the separate correlations. We illustrate our methods on two examples from a financial time series and genomic pathway analysis.

• Brown, Andrew; Jensen, Jimmy Kristian; Nikolova, Y.S.; Djurovic, Srdjan; Agartz, Ingrid & ALONSO, ANDRES SERVER [Show all 13 contributors for this article] (2012). Genetic variants affecting the neural processing of human facial expressions: evidence using a genome-wide functional imaging approach. Translational Psychiatry. doi: 10.1038/tp.2012.67. Full text in Research Archive
• Joshi, Himanshu; Nordgard, Silje H.; Frigessi, Arnoldo; Børresen-Dale, Anne-Lise & Kristensen, Vessela N. (2012). Overrepresentation of transcription factor families in the genesets underlying breast cancer subtypes. BMC Genomics. 13(199). doi: 10.1186/1471-2164-13-199. Full text in Research Archive
• Villani, Antonella; Frigessi, Arnoldo; Liljeros, Fredrik; Nordvik, Monika K. & Blasio, Birgitte Freiesleben De (2012). A Characterization of Internet Dating Network Structures among Nordic Men Who Have Sex with Men. PLOS ONE. 7(7). doi: 10.1371/journal.pone.0039717. Full text in Research Archive
• Serio, C di; Pellin, Danilo; ambrosi, alessandro; Glad, Ingrid Kristine & Frigessi, Arnoldo (2012). Biostatistics Meets Bioinformatics in Integrating Information from Highdimensional Heterogeneous Genomic Data: Two Examples from Rare Genetic Diseases and Infectious Diseases. In Biganzoli, E; Vellido, A; Ambrogi, F & Tagliaferri, R (Ed.), Computational Intelligence methods for bioinformatics and biostatistics. Springer. ISSN 9783642356865. p. 10–20. doi: 10.1007/978-3-642-35686-5_2. Full text in Research Archive
• LeBlanc, Marissa; Kulle, Bettina; Sundet, Kjetil Søren; Agartz, Ingrid; Melle, Ingrid & Djurovic, Srdjan [Show all 8 contributors for this article] (2012). Genome-wide study identifies PTPRO and WDR72 and FOXQ1-SUMO1P1 interaction associated with neurocognitive function. Journal of Psychiatric Research. ISSN 0022-3956. 46(2), p. 271–278. doi: 10.1016/j.jpsychires.2011.11.001. Full text in Research Archive
• Frigessi, Arnoldo; Løland, Anders; Pievatolo, Antonio & Ruggeri, Fabrizio (2011). Statistical rehabilitation of improper correlation matrices. Quantitative finance (Print). ISSN 1469-7688. 11(7), p. 1081–1090. doi: 10.1080/14697680903390118. Full text in Research Archive
• Ambrosi, Alessandro; Glad, Ingrid Kristine; Pellin, Danilo; Cattoglio, Claudia; Mavilio, Fulvio & serio, clelia di [Show all 7 contributors for this article] (2011). Estimated Comparative Integration Hotspots Identify Different Behaviors of Retroviral Gene Transfer Vectors. PLoS Computational Biology. ISSN 1553-734X. 7(12). doi: 10.1371/journal.pcbi.1002292. Full text in Research Archive
• Solvang, Hiroko Kato; Lingjærde, Ole Christian; Frigessi, Arnoldo; Børresen-Dale, Anne-Lise & Kristensen, Vessela N. (2011). Linear and non-linear dependencies between copy number aberrations and mRNA expression reveal distinct molecular pathways in breast cancer. BMC Bioinformatics. 12. doi: 10.1186/1471-2105-12-197. Full text in Research Archive Show summary

  Background Elucidating the exact relationship between gene copy number and expression would enable identification of regulatory mechanisms of abnormal gene expression and biological pathways of regulation. Most current approaches either depend on linear correlation or on nonparametric tests of association that are insensitive to the exact shape of the relationship. Based on knowledge of enzyme kinetics and gene regulation, we would expect the functional shape of the relationship to be gene dependent and to be related to the gene regulatory mechanisms involved. Here, we propose a statistical approach to investigate and distinguish between linear and nonlinear dependences between DNA copy number alteration and mRNA expression. Results We applied the proposed method to DNA copy numbers derived from Illumina 109 K SNP-CGH arrays (using the log R values) and expression data from Agilent 44 K mRNA arrays, focusing on commonly aberrated genomic loci in a collection of 102 breast tumors. Regression analysis was used to identify the type of relationship (linear or nonlinear), and subsequent pathway analysis revealed that genes displaying a linear relationship were overall associated with substantially different biological processes than genes displaying a nonlinear relationship. In the group of genes with a linear relationship, we found significant association to canonical pathways, including purine and pyrimidine metabolism (for both deletions and amplifications) as well as estrogen metabolism (linear amplification) and BRCA-related response to damage (linear deletion). In the group of genes displaying a nonlinear relationship, the top canonical pathways were specific pathways like PTEN and PI13K/AKT (nonlinear amplification) and Wnt(B) and IL-2 signalling (nonlinear deletion). Both amplifications and deletions pointed to the same affected pathways and identified cancer as the top significant disease and cell cycle, cell signaling and cellular development as significant networks. Conclusions This paper presents a novel approach to assessing the validity of the dependence of expression data on copy number data, and this approach may help in identifying the drivers of carcinogenesis.

• Jemtland, Rune; Holden, Marit; Reppe, Sjur; Olstad, Ole Kristoffer; Reinholt, Finn P. & Gautvik, Vigdis Teig [Show all 10 contributors for this article] (2011). Molecular Disease Map of Bone Characterizing the Postmenopausal Osteoporosis Phenotype. Journal of Bone and Mineral Research. ISSN 0884-0431. 26(8), p. 1793–1801. doi: 10.1002/jbmr.396. Full text in Research Archive
• Orskaug, Elisabeth; Scheel, Ida; Frigessi, Arnoldo; Guttorp, Peter; Haugen, Jan Erik & Tveito, Ole Einar [Show all 7 contributors for this article] (2011). Evaluation of a dynamic downscaling of precipitation over the Norwegian mainland. Tellus A: Dynamic Meteorology and Oceanography. ISSN 0280-6495. 63(4), p. 746–756. doi: 10.1111/j.1600-0870.2011.00525.x. Full text in Research Archive
• Gundersen, Sveinung; Kalaš, Matúš; Abul, Osman; Frigessi, Arnoldo; Hovig, Eivind & Sandve, Geir Kjetil (2011). Identifying elemental genomic track types and representing them uniformly. BMC Bioinformatics. 12. doi: 10.1186/1471-2105-12-494. Full text in Research Archive Show summary

  Background: With the recent advances and availability of various high-throughput sequencing technologies, data on many molecular aspects, such as gene regulation, chromatin dynamics, and the three-dimensional organization of DNA, are rapidly being generated in an increasing number of laboratories. The variation in biological context, and the increasingly dispersed mode of data generation, imply a need for precise, interoperable and flexible representations of genomic features through formats that are easy to parse. A host of alternative formats are currently available and in use, complicating analysis and tool development. The issue of whether and how the multitude of formats reflects varying underlying characteristics of data has to our knowledge not previously been systematically treated. Results: We here identify intrinsic distinctions between genomic features, and argue that the distinctions imply that a certain variation in the representation of features as genomic tracks is warranted. Four core informational properties of tracks are discussed: gaps, lengths, values and interconnections. From this we delineate fifteen generic track types. Based on the track type distinctions, we characterize major existing representational formats and find that the track types are not adequately supported by any single format. We also find, in contrast to the XML formats, that none of the existing tabular formats are conveniently extendable to support all track types. We thus propose two unified formats for track data, an improved XML format, BioXSD 1.1, and a new tabular format, GTrack 1.0. Conclusions: The defined track types are shown to capture relevant distinctions between genomic annotation tracks, resulting in varying representational needs and analysis possibilities. The proposed formats, GTrack 1.0 and BioXSD 1.1, cater to the identified track distinctions and emphasize preciseness, flexibility and parsing convenience.

• Zhao, Xi; Rødland, Einar Andreas; Sørlie, Therese; Naume, Bjørn; Langerød, Anita & Frigessi, Arnoldo [Show all 9 contributors for this article] (2011). Combining Gene Signatures Improves Prediction of Breast Cancer Survival. PLOS ONE. 6(3). doi: 10.1371/journal.pone.0017845. Full text in Research Archive

View all works in NVA

• Aalen, Odd O.; Frigessi, Arnoldo; Moger, Tron Anders; Scheel, Ida; Skovlund, Eva & Veierød, Marit Bragelien (2018). Statistiske metoder i medisin og helsefag. 2. utgave. Gyldendal Akademisk. ISBN 9788205507647. 351 p. Full text in Research Archive
• Frigessi, Arnoldo; Bühlmann, Peter; Glad, Ingrid Kristine; Langaas, Mette; Richardson, Sylvia Therese Lamblin & Vannucci, Marina (2016). Statistical Analysis for High-Dimensional Data. Springer. ISBN 9783319270975. 306 p. Full text in Research Archive
• Frigessi, Arnoldo; Buhlmann, A; Glad, I. K.; Langaas, M; Richardson, S & Vannicci, M (2016). Statistical analysis for high-dimensional data. The Abel Symposium 2014. Springer. ISBN 9783319270975. 360 p. Full text in Research Archive

View all works in NVA

• Eskeland, Gunnar; Glad, Ingrid Kristine & Frigessi, Arnoldo (2026). Europa må regulere KI | DN. Full text in Research Archive Show summary

  Høvdingene innen kunstig intelligens (KI) babler om trusler mot menneskeheten og ber om regulering, og da kreves skepsis i responsen (jf. Mogstad og Hardings kronikk i DN 2. oktober: «Hvorfor innovere når man kan regulere?»).Det de ber om, skal de ikke få.Ikke i Europa.

• Zuccato, Lorenzo; George, Anne-Marie; Vitelli, Valeria; Frigessi, Arnoldo; Giese, Martin & Scheel, Ida (2026). Bridging Statistical and Logical Perspectives on Inconsistency. Full text in Research Archive
• Haff, Ingrid Hobæk; Griesbauer, Elisabeth Maria; Czado, Claudia & Frigessi, Arnoldo (2026). Synthetic data with vine copulas – balancing utility and privacy. Full text in Research Archive
• Glad, Ingrid Kristine & Frigessi, Arnoldo (2025). KI = Statistikk2.0 ? Full text in Research Archive Show summary

  Foredrag i anledning den internasjonale statistikk-dagen

• Frigessi, Arnoldo (2025). Integrering av kunstig intelligens i pasientbehandling. Full text in Research Archive
• Frigessi, Arnoldo (2025). True data and their synthetic versions. Full text in Research Archive
• Frigessi, Arnoldo (2025). True data and their synthetic versions. Full text in Research Archive
• Frigessi, Arnoldo (2025). What can we expect from AI in the next 5 years? Full text in Research Archive
• Griesbauer, Elisabeth Maria; Czado, Claudia; Frigessi, Arnoldo & Haff, Ingrid Hobæk (2025). Synthetic data generation balancing privacy and utility, using vine copulas. Full text in Research Archive Show summary

  The availability of high-quality data has led to tremendous advances in science, technology and society at large, when analysed by means of statistical and machine learning (ML) methods. However, real-world data in many cases cannot be made public to the research community due to privacy restrictions. This impairs progress especially in bio-medical research. Synthetic data can substitute the sensitive real data, as long as they do not disclose private aspects. This has proven to be successful in training downstream ML applications. We propose TVineSynth, a vine copula based synthetic tabular data generator. TVineSynth is designed to balance privacy and utility, using the vine tree structure and its truncation to do the trade-off. Contrary to synthetic data generators that achieve differential privacy (DP) by globally adding noise, TVineSynth performs a controlled approximation of the estimated data generating distribution. Because of this it does not suffer from poor utility of the resulting synthetic data for downstream prediction tasks. TVineSynth introduces a targeted bias into the vine copula model. Combined with the specific tree structure of the vine, this causes the model to zero out privacy-leaking dependencies while relying on those that are beneficial for utility. We theoretically justify how the construction of TVineSynth ensures privacy. When compared to competitor models, with and without DP, TVineSynth achieves a superior privacy-utility balance.

• Griesbauer, Elisabeth Maria; Czado, Claudia; Frigessi, Arnoldo & Haff, Ingrid Hobæk (2025). Poster: TVineSynth: A Truncated C-Vine Copula Generator of Synthetic Tabular Data to Balance Privacy and Utility. Full text in Research Archive Show summary

  We propose TVineSynth, a vine copula based synthetic tabular data generator, which is designed to balance privacy and utility, using the vine tree structure and its truncation to do the trade-off. Contrary to synthetic data generators that achieve DP by globally adding noise, TVineSynth performs a controlled approximation of the estimated data generating distribution, so that it does not suffer from poor utility of the resulting synthetic data for downstream prediction tasks. TVineSynth introduces a targeted bias into the vine copula model that, combined with the specific tree structure of the vine, causes the model to zero out privacy-leaking dependencies while relying on those that are beneficial for utility. Privacy is here measured with membership (MIA) and attribute inference attacks (AIA). Further, we theoretically justify how the construction of TVineSynth ensures AIA privacy under a natural privacy measure for continuous sensitive attributes. When compared to competitor models, with and without DP, on simulated and on real-world data, TVineSynth achieves a superior privacy-utility balance.

• Griesbauer, Elisabeth Maria; Czado, Claudia; Frigessi, Arnoldo & Haff, Ingrid Hobæk (2025). Poster: TVineSynth: A Truncated C-Vine Copula Generator of Synthetic Tabular Data to Balance Privacy and Utility. Full text in Research Archive Show summary

  We propose TVineSynth, a vine copula based synthetic tabular data generator, which is designed to balance privacy and utility, using the vine tree structure and its truncation to do the trade-off. Contrary to synthetic data generators that achieve DP by globally adding noise, TVineSynth performs a controlled approximation of the estimated data generating distribution, so that it does not suffer from poor utility of the resulting synthetic data for downstream prediction tasks. TVineSynth introduces a targeted bias into the vine copula model that, combined with the specific tree structure of the vine, causes the model to zero out privacy-leaking dependencies while relying on those that are beneficial for utility. Privacy is here measured with membership (MIA) and attribute inference attacks (AIA). Further, we theoretically justify how the construction of TVineSynth ensures AIA privacy under a natural privacy measure for continuous sensitive attributes. When compared to competitor models, with and without DP, on simulated and on real-world data, TVineSynth achieves a superior privacy-utility balance.

• Glad, Ingrid Kristine & Frigessi, Arnoldo (2025). Nysgjerrige på: neste generasjon maskinlæring - Nysgjerrige Norge | Acast. Full text in Research Archive Show summary

  Listen to Nysgjerrige på: neste generasjon maskinlæring from Nysgjerrige Norge. Nysgjerrige Norge - Episode 3: Et dypdykk i kunnskapsdrevet maskinlæring med IntegreatI denne episoden tar Kristopher turen til Nils Henrik Abels hus på Blindern i Oslo for å besøke Integreat Senter for Fremragende Forskning. Vi møter professorene og ekteparet Arnoldo Frigessi og Ingrid Glad, som leder det nyoppstartede senteret. De deler visjoner om fremtidens algoritmer, hvor kunnskap og etikk spiller en nøkkelrolle i utviklingen av kunstig intelligens (KI).Nysgjerrige Norge er en serie fra Norges forskningsråd. Sentre for Fremragende Forskning er en støtteordning til landets fremste vitenskapelige miljøer. Du kan lese mer om støtteordningen her. Serien er produsert av Moose Media. Programleder: Kristopher Schau. Musikk: The Dogs. Abonnér på Nysgjerrige Norge og møt flere av landets fremste vitenskapspersonligheter.

• Storvik, Geir Olve; Engebretsen, Solveig; Blasio, Birgitte Freiesleben De & Frigessi, Arnoldo (2025). Flaws in the Article “Nearly Instantaneous Time-Varying Reproduction Number for Contagious Diseases—a Direct Approach Based on Nonlinear Regression". Journal of Computational Biology. ISSN 1066-5277. 32(8), p. 813–818. doi: https:/www.liebertpub.com/doi/full/10.1177/15578666251360613. Full text in Research Archive
• Engebretsen, Solveig; Thorarinsdottir, Thordis Linda; Palomares, Alfonso Diz-Lois; Storvik, Geir Olve; Frigessi, Arnoldo & Blasio, Birgitte Freiesleben De (2025). Contribution to the Discussion of 'Some statistical aspects of the Covid-19 response' by Wood et al. Journal of the Royal Statistical Society, Series A (Statistics in Society). ISSN 0964-1998. 189(1), p. 71–72. doi: 10.1093/jrsssa/qnaf080. Full text in Research Archive
• Griesbauer, Elisabeth Maria; Czado, Claudia; Frigessi, Arnoldo & Haff, Ingrid Hobæk (2024). Synthetic data generation balancing privacy and utility, using vine copulas. doi: https:/nva.sikt.no/registration/0199812d9067-f1497889-e36d-4995-8686-7c68d1e99cdb/edit. Full text in Research Archive Show summary

  The availability of diverse, high-quality data has led to tremendous advances in science, technology and society at large, when analysed by means of statistical and machine learning (ML) methods. However, real-world data, in many cases, cannot be made public to the research community due to privacy restrictions, obstructing progress, especially in bio- medical research. Synthetic data can substitute the sensitive real data, and as long as they do not disclose private aspects. This has proven to be successful in training downstream ML applications. We propose TVineSynth, a vine copula based synthetic tabular data generator, which is designed to balance privacy and utility, using the vine tree structure and its truncation to do the trade-off. Contrary to synthetic data generators that achieve differential privacy (DP) by globally adding noise, TVineSynth performs a controlled approximation of the estimated data generating distribution, so that it does not suffer from poor utility of the resulting synthetic data for downstream prediction tasks. TVineSynth introduces a targeted bias into the vine copula model that, combined with the specific tree structure of the vine, causes the model to zero out privacy-leaking dependencies while relying on those that are beneficial for utility. Privacy is here measured with membership (MIA) and attribute inference attacks (AIA). Further, we theoretically justify how the construction of TVineSynth ensures AIA privacy under a natural privacy measure for continuous sensitive attributes. When compa red to competitor models, with and without DP, on simulated and on real-world data, TVineSynth achieves a superior privacy-utility balance.

• Frigessi, Arnoldo (2024). Om kunstig intelligens - NRK Nyhetsmorgen (2. januar 2024). [TV]. NRK. Full text in Research Archive
• Haff, Ingrid Hobæk; Griesbauer, Elisabeth Maria; Frigessi, Arnoldo & Czado, Claudia (2024). Synthetic data with vine copulas – balancing utility and privacy. Full text in Research Archive
• Haff, Ingrid Hobæk; Griesbauer, Elisabeth Maria; Frigessi, Arnoldo & Czado, Claudia (2024). Synthetic data with vine copulas – balancing utility and privacy. Full text in Research Archive
• Frigessi, Arnoldo (2024). Supporting the AI revolution. Full text in Research Archive
• Frigessi, Arnoldo (2024). Kunstig intelligens for anestesiforskning og praksis. Full text in Research Archive
• Zucknick, Manuela; Frigessi, Arnoldo & Smajdor, Anna (2024). Autonomy and decision-making in the era of medical AI: the red pill or the blue? Full text in Research Archive
• Pérez, Tomás Fernando Varnet; Øvergaard, Kristin Romvig; Frigessi, Arnoldo & Biele, Guido (2024). Design-aware imputation in a target trial of ADHD pharmaceutical treatment on Norwegian national test scores: the case of simultaneous outcomes. Full text in Research Archive
• Cavalieri, Stefano; Vener, Claudia; Leblanc, Marissa Erin; Lopez-Perez, Laura; Fico, Giuseppe & Resteghini, Carlo [Show all 23 contributors for this article] (2023). A multicenter randomized trial for quality of life evaluation by non-invasive intelligent tools during post-curative treatment follow-up for head and neck cancer: Clinical study protocol. Frontiers in Oncology. 13, p. 1–16. doi: 10.3389/fonc.2023.1048593. Full text in Research Archive Show summary

  Patients surviving head and neck cancer (HNC) suffer from high physical, psychological, and socioeconomic burdens. Achieving cancer-free survival with an optimal quality of life (QoL) is the primary goal for HNC patient management. So, maintaining lifelong surveillance is critical. An ambitious goal would be to carry this out through the advanced analysis of environmental, emotional, and behavioral data unobtrusively collected from mobile devices. The aim of this clinical trial is to reduce, with non-invasive tools (i.e., patients’ mobile devices), the proportion of HNC survivors (i.e., having completed their curative treatment from 3 months to 10 years) experiencing a clinically relevant reduction in QoL during follow-up. The Big Data for Quality of Life (BD4QoL) study is an international, multicenter, randomized (2:1), open-label trial. The primary endpoint is a clinically relevant global health-related EORTC QLQ-C30 QoL deterioration (decrease ≥10 points) at any point during 24 months post-treatment follow-up. The target sample size is 420 patients. Patients will be randomized to be followed up using the BD4QoL platform or per standard clinical practice. The BD4QoL platform includes a set of services to allow patients monitoring and empowerment through two main tools: a mobile application installed on participants’ smartphones, that includes a chatbot for e-coaching, and the Point of Care dashboard, to let the investigators manage patients data. In both arms, participants will be asked to complete QoL questionnaires at study entry and once every 6 months, and will undergo post-treatment follow up as per clinical practice. Patients randomized to the intervention arm (n=280) will receive access to the BD4QoL platform, those in the control arm (n=140) will not. Eligibility criteria include completing curative treatments for non-metastatic HNC and the use of an Android-based smartphone. Patients undergoing active treatments or with synchronous cancers are excluded.

• Frigessi, Arnoldo (2023). Hjernen i sentrum: Kunstig intelligens Hva er best; kunstig eller ekte intelligens? Full text in Research Archive
• Haff, Ingrid Hobæk; Griesbauer, Elisabeth Maria; Frigessi, Arnoldo & Czado, Claudia (2023). Synthetic data with vine-copulas - balancing utility and privacy. Full text in Research Archive
• Frigessi, Arnoldo (2023). From limited patient data, to high frequency synthetic data, to the differential equation of a breast tumour growth. Full text in Research Archive
• Frigessi, Arnoldo (2023). The AI algorithm that you are using has its own agenda: you might want to know it. Full text in Research Archive
• Frigessi, Arnoldo; Smajdor, Anna & Zucknick, Manuela (2023). The Importance of Reward Functions in Machine Learning Algorithms and Why These Should be Regulated. Full text in Research Archive Show summary

  Session III: AI in Practice Presentation: The Importance of Reward Functions in Machine Learning Algorithms and Why These Should be Regulated presented by Frigessi

• Frigessi, Arnoldo (2023). Pair copula constructions in machine learning. Full text in Research Archive
• Frigessi, Arnoldo (2023). Kunstig intelligens, algoritmer i klinisk praksis, hvor er vi, hva skjer nå, muligheter og utfordringer. Full text in Research Archive
• Frigessi, Arnoldo (2023). The AI ecosystem. Full text in Research Archive
• Frigessi, Arnoldo (2023). AI i medisin og klinikk. Full text in Research Archive
• Pérez, Tomás Fernando Varnet; Biele, Guido; Frigessi, Arnoldo & Øvergaard, Kristin Romvig (2023). Synthetic data for unidentified Bayesian multiple bias models an application to ADHD interventions for academic achievement. Full text in Research Archive
• Andreassen, Tor Arne; Frigessi, Arnoldo & Spurkland, Anne (2022). Omikron vil neppe smitte alle. Men en eller annen variant, vil nesten alle bli smittet av. [Journal]. Aftenposten. Full text in Research Archive
• Frigessi, Arnoldo (2022). Probabilistic preference learning from incomplete rank data. Full text in Research Archive
• Frigessi, Arnoldo (2022). A Bayesian two-way latent structure clustering model for multiple data integration, with an application to subtyping breast cancer. Full text in Research Archive
• Frigessi, Arnoldo (2022). A Bayesian two-way latent structure clustering model for multiple data integration, with an application to subtyping breast cancer. Full text in Research Archive
• Frigessi, Arnoldo (2021). Data Science for the covid-19 emergency. Full text in Research Archive
• Stoltenberg, Camilla; Aavitsland, Preben; Rø, Gunnar Øyvind Isaksson; Blasio, Birgitte Freiesleben de; Rattalma, Arnoldo Frigessi Di & Engebretsen, Solveig (2021). Aftenposten bommer om prognoser. Igjen og igjen. Full text in Research Archive
• Frigessi, Arnoldo (2021). Mathematical modelling of the Sars-CoV-2 pandemic in Norway. Full text in Research Archive
• Frigessi, Arnoldo (2021). Data science in an emergency. Full text in Research Archive
• Frigessi, Arnoldo (2021). Models and predictions during the Covid-19 emergency in Norway. Full text in Research Archive
• Frigessi, Arnoldo (2021). Sverdrup prisen 2021. Full text in Research Archive
• Frigessi, Arnoldo (2021). Data science in emergency. Full text in Research Archive
• Frigessi, Arnoldo (2021). The Norwegian regional stochastic model of the Covid-19 epidemics. Full text in Research Archive
• Frigessi, Arnoldo (2021). Math&Stats modelling of breast cancer therapy. Full text in Research Archive
• Frigessi, Arnoldo (2021). Covid-19 modelling and AI. Full text in Research Archive
• Frigessi, Arnoldo (2021). Debatt i NRK radio P2 om utnyttelse av norske helsedata. [Radio]. NRK P2. Full text in Research Archive
• Stoltenberg, Camilla; Rattalma, Arnoldo Frigessi Di; Engebretsen, Solveig; Sandberg, Hallvard; Johansen, Per Anders & Gjuvsland, Elin Ruhlin (2021). Paneldebatt: HVA KAN VI LÆRE AV RUNDEN MED COVID: HVORDAN HÅNDTERER VI USIKKERHET OG ÅPENHET? [Internet]. Arendalsuka. Full text in Research Archive
• Løland, Anders & Rattalma, Arnoldo Frigessi Di (2020). Episode 2: Om den essensielle usikkerheten. Med Arnoldo Frigessi. [Internet]. Sannsynligvis VIKTIG (podkast). Full text in Research Archive
• Løland, Anders & Rattalma, Arnoldo Frigessi Di (2020). Episode 3: Hvorfor skal forskning være en del av beredskapen vår? Med Arnoldo Frigessi. [Internet]. Sannsynligvis VIKTIG (podkast). Full text in Research Archive
• Rattalma, Arnoldo Frigessi Di (2020). The Norwegian Covid-19 Model: a spatial modelling of the early phase epidemics. Full text in Research Archive
• Rattalma, Arnoldo Frigessi Di (2020). Spatial modelling of early-phase COVID-19 epidemic in Norway. Full text in Research Archive
• Rattalma, Arnoldo Frigessi Di (2020). In silico modelling of breast tumours for personalised therapy. Full text in Research Archive
• Rattalma, Arnoldo Frigessi Di; Köhn-Luque, Alvaro; Kristensen, Vessela N.; Gørbitz, Carl Henrik & Engebråten, Olav (2020). Kan statistikk og matematikk redde like mange liv som nye medisiner? Aftenposten (morgenutg. : trykt utg.). ISSN 0804-3116. Full text in Research Archive
• Lai, Xiaoran; Luque, Alvaro Kohn; Frigessi, Arnoldo & Geier, Oliver (2017). Breast MRI as a building block for patient-specific simulations of drug effect. Full text in Research Archive
• Frigessi, Arnoldo (2016). Big Data Big Insight. Full text in Research Archive
• Frigessi, Arnoldo (2016). Big Health data. Full text in Research Archive
• Frigessi, Arnoldo (2016). A high-dimensional tribute to Juni Palmgren. Full text in Research Archive
• Frigessi, Arnoldo (2016). Value from sharing data. Full text in Research Archive
• Frigessi, Arnoldo (2016). Big health data: fantastiske muligheter og utfordringer. Full text in Research Archive
• Frigessi, Arnoldo (2016). Bayesian Inference for Ranks. Full text in Research Archive
• Frigessi, Arnoldo (2016). My data science. Full text in Research Archive
• Frigessi, Arnoldo (2016). Discovery and prediction of change. Full text in Research Archive
• Frigessi, Arnoldo (2016). From big data to big insight. Full text in Research Archive
• Frigessi, Arnoldo (2016). The power of statistical science: from data to discovery and value. Full text in Research Archive
• Frigessi, Arnoldo; Vitelli, Valeria; Crispino, Marta & Arjas, Elja (2016). Recommendations from non-transitive pairwise comparisons. Full text in Research Archive
• Frigessi, Arnoldo (2016). The value of uncertainty. Full text in Research Archive
• Frigessi, Arnoldo & Løland, Anders (2016). Forskning på statistiske modeller kan gi store gevinster. [Journal]. Forskningsnytt 2015. Full text in Research Archive
• Haug, Ola; Bolin, David; Frigessi, Arnoldo; Guttorp, Peter; Orskaug, Elisabeth & Scheel, Ida [Show all 7 contributors for this article] (2016). Modelling and predicting residential water damage insurance claims via a calibrated dynamical downscaling. Full text in Research Archive
• Frigessi, Arnoldo; Buhlmann, Peter; Glad, Ingrid Kristine; Langaas, Mette; Richardson, Sylvia Therese Lamblin & Vannucci, Marina (2016). Abel Symposia: Preface. Abel Symposia. ISSN 2193-2808. 11, p. vii–x. Full text in Research Archive
• Frigessi, Arnoldo (2016). The data rich society of the future. Full text in Research Archive
• Frigessi, Arnoldo (2016). BigInsight and Skatteetatten. Full text in Research Archive
• Frigessi, Arnoldo (2015). Bayesian Mallows model for preference learning. Full text in Research Archive
• Frigessi, Arnoldo (2015). Apollon lansering. Full text in Research Archive
• Frigessi, Arnoldo (2015). Meta analysis with Mallows rank model. Full text in Research Archive
• Frigessi, Arnoldo (2015). Pan Cancer models. Full text in Research Archive
• Frigessi, Arnoldo (2015). Preference learning. Full text in Research Archive
• Haff, Ingrid Hobæk; Frigessi, Arnoldo & Maraun, Douglas (2015). Hvor godt klarer regionale klimamodeller å gjenskape den romlige avhengigheten i nedbør? Full text in Research Archive
• Haff, Ingrid Hobæk; Frigessi, Arnoldo & Maraun, Douglas (2015). How well do regional climate models simulate the spatial dependence of precipitation? An application of pair-copula constructions. Full text in Research Archive
• Frigessi, Arnoldo (2015). Her er likheten mellom kreft og skattejuks. [Journal]. Apollon. Full text in Research Archive
• Frigessi, Arnoldo (2015). Innovasjon vs. personvern. [Journal]. ComputerWorld. Full text in Research Archive
• Frigessi, Arnoldo (2015). De svakes ridder. [Journal]. Apollon. Full text in Research Archive
• Haug, Ola; Frigessi, Arnoldo; Scheel, Ida & Guttorp, Peter (2015). Modelling and predicting residential water damage insurance claims in a climate change perspective. Full text in Research Archive
• Frigessi, Arnoldo (2015). The future of biostatistics and the Oslo Center for Biostatistics and Epidemiology. Full text in Research Archive
• Frigessi, Arnoldo (2015). kurs STK1110, om big data og fremtiden i statistikk. Full text in Research Archive
• Frigessi, Arnoldo (2015). Ledermøte: presentasjon av BigInsight. Full text in Research Archive
• Frigessi, Arnoldo (2015). Stabmøte "What is happening in biostatistics". Full text in Research Archive
• Frigessi, Arnoldo (2015). SFI Big Insight - Norsk Regnesentral/UiO. Full text in Research Archive
• Frigessi, Arnoldo (2014). Bayesian inference for ranks. Lecture Notes in Computer Science (LNCS). ISSN 0302-9743. 8819, p. XIII–XIV. doi: 10.1007/978-3-319-12571-8. Full text in Research Archive
• Haug, Ola; Scheel, Ida; Orskaug, Elisabeth; Frigessi, Arnoldo; Guttorp, Peter & Ferkingstad, Egil (2014). Vulnerability models for water damage insurance claims - predictions of future losses in a climate change perspective. Full text in Research Archive
• Haug, Ola; Orskaug, Elisabeth; Scheel, Ida; Frigessi, Arnoldo; Maraun, Douglas & Guttorp, Peter (2012). Evaluation and calibration of dynamically downscaled precipitation over Norwegian mainland. Full text in Research Archive
• Tramm, T; Mohammed, H; Myhre, S; Alsner, J; Borresen-Dale, AL & Sorlie, T [Show all 8 contributors for this article] (2012). A Seven-gene Signature Predicting Benefit of Postmastectomy Radiotherapy in High Risk Breast Cancer. European Journal of Cancer. ISSN 0959-8049. 48, p. S181–S181. Full text in Research Archive
• Haug, Ola; Orskaug, Elisabeth; Scheel, Ida; Frigessi, Arnoldo; Maraun, Douglas & Guttorp, Peter (2012). Evaluation and Calibration of Dynamically Downscaled Precipitation over Norwegian Mainland. Full text in Research Archive
• Scheel, Ida; Haug, Ola; Orskaug, Elisabeth; Frigessi, Arnoldo & Guttorp, Peter (2012). Evaluating and Calibrating Dynamically Downscaled Precipitation Using the Doksum Shift Function. Full text in Research Archive
• Holden, Lars; Frigessi, Arnoldo & Mysterud, Rønnaug Sægrov (2012). Statistics for innovation, presentasjon av en SFI. Full text in Research Archive
• Frigessi, Arnoldo; Holden, Lars & Aalen, Odd O. (2011). Klima og statistikk. Full text in Research Archive
• Orskaug, Elisabeth; Scheel, Ida; Frigessi, Arnoldo; Guttorp, Peter; Haugen, Jan Erik & Tveito, Ole Einar [Show all 7 contributors for this article] (2011). Evaluation of a dynamic downscaling of Norwegian precipitation. Full text in Research Archive
• Haug, Ola; Orskaug, Elisabeth; Scheel, Ida; Frigessi, Arnoldo; Guttorp, Peter & Maraun, Douglas (2011). Calibrating dynamically downscaled precipitation using the Doksum shift function. Full text in Research Archive
• Haug, Ola; Orskaug, Elisabeth; Scheel, Ida; Frigessi, Arnoldo; Guttorp, Peter & Maraun, Douglas (2011). Calibrating dynamically down-scaled precipitation using the Doksum shift function. Full text in Research Archive
• Pérez, Tomás Fernando Varnet; Biele, Guido; Frigessi, Arnoldo & Øvergaard, Kristin Romvig (2026). From real-world data to causal claims: Target trial emulation, sensitivity and secondary analyses of ADHD interventions’ effects on academic achievement. Universitetet i Oslo. ISSN 9788234809132. Full text in Research Archive Show summary

  The present thesis is part of the project led by Guido Biele, 'Effects of pharmacological treatment & special education on school performance of children with ADHD', and thus aims to answer two main questions: (i) What is the long-term effect of pharmacological treatment on academic outcomes in children diagnosed with ADHD? and (ii) What is the long-term effect of special education provision on academic outcomes in children diagnosed with ADHD? The problem is thus one of causal inference based on observational data sourced from Norwegian national registries and the Norwegian Mother, Father and Child Cohort Study (MoBa). The use of 'real-world data' is particularly challenging as it can provide relevant evidence only when coupled with strong assumptions. Thus, the present work focuses on the variety of analyses and approaches that can strengthen our inferences by weakening or complementing our reliance on these assumptions. The thesis consists of an introduction and three articles. The introduction examines key epistemological and substantive issues relevant to this work, particularly the field of causal inference and the nature of ADHD. Paper I answers the first question via a target trial emulation of a pretest-posttest control group design, leveraging negative control analysis, proximal causal inference, a flexible machine learning semi-parametric model and secondary analyses on different subsets of the population. Paper II provides an extension of an existing sensitivity analysis for unobserved confounding to allow the use of bias prior distributions in a Bayesian model, applying this method to answer both questions while weakening the no unobserved confounding assumption. Paper III introduces an approach that leverages multiversal and meta-analytic tools to reconcile conflicting results between studies with different designs, as well as replicating a previous study from the literature in the Norwegian population. The thesis concludes with a review of treatment effects including an extended background on the main hypotheses about the effect of ADHD interventions, an extended discussion of these articles and future research directions.

• Griesbauer, Elisabeth Maria & Frigessi, Arnoldo (2026). Vine Copulas in Machine Learning. Universitetet i Oslo. ISSN 9788234809064. Full text in Research Archive Show summary

  The availability of high-quality data and the reliable quantification of uncertainty about conclusions drawn from that data, are two key factors for scientific progress in biomedical research and for clinical decision making. However, privacy concerns about patient-level data prohibit (easy) sharing of data. This hinders broad research collaborations, the re-production of existing study results and the data serving as benchmarks for methodological developments in bio-statistics and machine learning (ML), and thus slows down research progress in bio-statistics and ML, and in medicine in general. In medical research and practice, it is ultimately patients’ well-being and lives that are at stake. Therefore, research findings cannot be ethically applied to clinical practice and decision making, if they are not accompanied by appropriate quantification of uncertainty that can be taken into account. However, uncertainty quantification can be computationally costly, hindering its application in biomedical research. ML is a toolbox with solutions to both of these challenges. Private generative modeling produces synthetic data based on the sensitive real data. The synthetic data can be published and used in place of the real data, contributing to preserving individuals’ privacy. Variational inference efficiently approximates and quantifies uncertainty in complex settings. Depending on the choice of approximate posterior, complex unobserved dependencies can be modeled at a moderate computational cost. Vine copulas are a statistical model that flexibly models complex dependence in a fine-grained fashion and has been widely applied in medical research. This thesis develops vine copula based methods for ML, namely for private generative modeling for synthetic data generation, and variational inference for uncertainty quantification. More specifically, the structure of vine copulas is exploited to balance privacy and utility in synthetic data generation in the first part, and to guide the flexibility in the uncertainty estimate in the second part.

• Krogstie, John; Storsul, Tanja; Frigessi, Arnoldo; Lyngstad, Cathrine Pihl; Andreassen, Eirik & Martinkenaite, Ieva [Show all 13 contributors for this article] (2023). Overordnede tiltak - Andre notat fra rådgivende ekspertgruppe for KI-satsingen. Norges forskningsråd. Full text in Research Archive
• Storsul, Tanja; Rettberg, Jill Walker; Brandtzæg, Petter Bae; Frigessi, Arnoldo; Strumke, Inga & Mahler, Tobias [Show all 12 contributors for this article] (2023). 1. notat fra rådgivende ekspertgruppe for KI-satsingen: Nåsituasjon og forventninger. Forskningsrådet. Full text in Research Archive
• Blasio, Birgitte Freiesleben de; Ruscio, Francesco Di; Rø, Gunnar Øyvind Isaksson; Engebretsen, Solveig; Frigessi, Arnoldo & Palomares, Alfonso Diz-Lois [Show all 14 contributors for this article] (2020). Situational awareness and forecasting - FHI COVID-19 modelling team. Folkehelseinstituttet. Full text in Research Archive
• Vitelli, Valeria; Sørensen, Øystein; Frigessi, Arnoldo & Arjas, Elja (2018). Probabilistic preference learning with the Mallows rank model. Journal of Machine Learning Research. Full text in Research Archive
• Haff, Ingrid Hobæk; Frigessi, Arnoldo & Aas, Kjersti (2012). Pair-copula constructions - an inferential perspective. Universitetet i Oslo. Full text in Research Archive
• Scheel, Ida; Ferkingstad, Egil; Frigessi, Arnoldo; Haug, Ola; Hinnerichsen, Mikkel & Meze-Hausken, Elisabeth (2011). A Bayesian hierarchical model with spatial variable selection: the effect of weather on insurance claims. Derivation of distributions and MCMC sampling schemes. Universitetet i Oslo. Full text in Research Archive Show summary

  Climate change will affect the insurance industry. We develop a Bayesian hierarchical statistical approach to explain and predict insurance losses due to weather events at a local geographical scale. The number of weather-related insurance claims is modelled combining generalized linear models with spatially smoothed variable selection. Using Gibbs sampling and reversible jump MCMC, the model is fitted on daily weather and insurance data from each of the 319 municipalities of southern and central Norway for the period 1997-2006. Out-of-sample predictions from the model are very good. Our results show interesting regional patterns in the impact of different weather covariates. In addition to being useful for insurance pricing, our model can be used for short-term predictions based on weather forecasts and long-term predictions based on downscaled climate models.

View all works in NVA