Sophia Anna Wilhelm

• Signorelli, Letizia; Korchynska, Solomiia; Ortiz, Cantin; Arena, Alessandro; Wilhelm, Sophia Anna & Boccara, Charlotte (2026). Nyx: a flexible framework for sleep scoring across species, lifespan and modalities. bioRxiv - the preprint server for biology. doi: 10.64898/2026.07.24.740558. Fulltekst i vitenarkiv Vis sammendrag

  Abstract Sleep is a key factor in almost every physiological process, yet scoring tools lag behind innovations in basic to clinical research. New automated algorithms are trained on narrow datasets, typically healthy adult mouse or human, and fail to generalise, leaving labs to fall back on slow, laborious, and often poorly reproducible manual scoring. Here we introduce Nyx, an unsupervised framework built on principal component analysis that flexibly clusters epochs into sleep states, akin to how spike-sorting algorithms cluster single-neuron activity. Validated on about 8,000 hours spanning 19 independent datasets, it generalises across species, lifespan, recording modalities, and altered sleep architectures, matching human inter-scorer agreement. Beyond this, Nyx opens new ground for modern sleep research: classifying NREM subtypes in rodents, revealing states that cut across canonical boundaries in non-mammalian species, and enabling real-time, closed-loop manipulation. Distributed as an open-source package with a graphical user interface, Nyx offers an accessible, explainable framework built to scale with the pace of innovation itself.

Se alle arbeider i NVA

• Signorelli, Letizia; Korchynska, Solomiia; Ortiz, Cantin; Arena, Alessandro; Wilhelm, Sophia Anna & Boccara, Charlotte (2026). Sleep scoring with Nyx: a generalisable framework across species, ages and recording Methods. Fulltekst i vitenarkiv Vis sammendrag

  Background and aims: Sleep scoring is essential for the characterisation of sleep architecture and supports most analyses in sleep research, yet it remains a time-consuming task. While automated sleep scoring has the potential to reduce manual workload and inter-rater variability, its adoption remains limited. Most existing tools are tailored to specific datasets, reducing their generalisability, and function as black-box algorithms, limiting interpretability and undermining researchers' trust. To address these challenges, we present Nyx, a generalisable, explainable, and accessible sleep scoring framework, based on Principal Component Analysis (PCA) for transparent feature extraction and flexible stage clustering. We demonstrate its application across species, lifespan, and recording modalities, and release it as an open-source tool with a graphical user interface (GUI) for non-technical users. Methods: Nyx was first validated on rodent polysomnographic (PSG) data from 17 datasets (10 baseline and 7 experimental manipulations; rats and mice; ~7,937 h of recordings). Then generalisation was evaluated across three dimensions: species (rodents, humans and additional species), developmental stage (juvenile, adolescent and adult), and recording modality, including acquisition setup (wired, wireless and telemetric) and signal combinations (electroencephalography (EEG) with electromyography (EMG), EEG with video or accelerometry, and EEG-only). Agreement with human scorers and established automated tools was quantified using Cohen's κ and F1 score. Preliminary evaluation of real-time sleep classification was also conducted to explore the framework's potential for closed-loop applications. Results: Nyx achieved performance comparable to inter-rater human agreement across all rodent baseline and experimental datasets (κ ≥ 0.80, F1 ≥ 0.85). The framework generalised across species and developmental stages, accurately differentiating human sleep stages (4-stage: κ ≥ 0.70, F1 ≥ 0.80; 5-stage: κ ≥ 0.65, F1 ≥ 0.75), while maintaining consistent performance across age groups. Real-time classification performance indicates suitability for closed-loop experimental applications. Conclusions: Nyx provides a reproducible, generalisable and accessible framework for sleep scoring across species, developmental stages and recording modalities. Its transparent architecture based on PCA addresses a key barrier to the adoption of automated scoring approaches. By enabling consistent analysis across diverse datasets and acquisition methods, Nyx facilitates more standardised sleep research and the exploration of cross-species, lifespan and cross-modality insights.

Se alle arbeider i NVA