Biology subjects

Westergaard, D.

Publications and source records attributed to Westergaard, D..

4 recordsLinked to original sources

bioRxiv2026

Menstrual cycle phase length variation is associated with daily symptom burden

Menstrual symptoms vary across the cycle, yet most research assumes a normative 28-day cycle with fixed phase durations, obscuring the physiological relevance of natural cycle variation. Using the mcPHASES dataset, we characterised cycle and phase length variation across 96 menstrual cycles from 37 participants, with ovulation timing estimated from daily urinary luteinizing hormone measurements using a Bayesian hierarchical model, and examined associations with daily symptoms in a subset of 64 cycles from 35 participants with complete symptom data. Twelve physical, mental, and behavioural symptom domains were modelled using Bayesian ordinal regression, with posterior uncertainty in phase-length predictors propagated via a measurement error framework. Total cycle length was not associated with daily symptom burden, except sleep disturbances. By contrast, phase length decomposition revealed systematic associations across multiple domains: longer menstrual phase length was broadly associated with greater symptom intensity spanning physical, gastrointestinal, affective, and sleep domains; longer luteal phase duration was associated with greater fatigue and more frequent headaches, but lower sore breast intensity and lower stress; and longer follicular phase duration and later ovulation were each associated with greater sore breast intensity and more frequent mood swings. These associations require knowledge of actual ovulation timing and cannot be recovered from cycle length alone, indicating that the common assumption of a fixed 14-day luteal phase introduces systematic misclassification of hormonal exposure. Daily symptom intensity was also predominantly person-specific, with cycle phase explaining little of the between-person variance across most symptoms. These findings indicate that calendar-based phase assignment is insufficient for research and clinical assessment of hormone-sensitive conditions, and that person-specific baselines, rather than population-level phase averages, are needed for clinically meaningful symptom monitoring.

physiology↗

bioRxiv2025

Population-scale analysis of inheritance patterns across 858,635 individuals reveals recent historical migration patterns across the North Sea from the Middle Ages to the Industrial Revolution

The North Seas historical migrations have impacted the genetic structure of its neighbouring populations. We analysed haplotype sharing among 858,635 modern individuals from Denmark and Britain to infer migration history from the Middle Ages to the modern day. We estimated the genetic relationship among 370,259 Danes using Danish healthcare registries and validated those with retrievable pedigree relationships from a national family registry. We also compared their haplotype sharing with 488,376 British individuals from the UK Biobank. Our analysis revealed the fine-grained population genetic history of Denmark and identified distinct coastal and island communities with a history of genetic isolation and bottlenecks. We observed a significant population decline in Jutland compared to Zealand from the late medieval period to the start of early modern period, accompanied by migration from Jutland to Zealand, corresponding to historical evidence. We identified two major haplotype sharing patterns between Denmark and Great Britain: (1) early coast-to-coast connections between South Jutland and eastern England, likely driven by migration, settlement, continuous trade and the movement of people across the North Sea from the Middle Ages through early modern times, and (2) later city-to-city connections such as those between London and Kobenhavn, likely shaped by the reorientation of economic and trading centres in Britain and Denmark, and the increased shipping from early modern times towards the Industrial Revolution. Further comparisons using other North Sea countries showed both shared and unique histories of genetic exchange with Denmark and Britain. Our study provides novel genetic evidence of migration across the North Sea from the Middle Ages to the Industrial Revolution. It highlights the power of nationwide biobanks in reconstructing fine-scale historical population movements among closely related populations.

genetics↗

bioRxiv2021

Deep integrative models for large-scale human genomics

Polygenic risk scores (PRSs) are expected to play a critical role in achieving precision medicine. Currently, PRS predictors are generally based on linear models using summary statistics, and more recently individual-level data. However, these predictors mainly capture additive relationships and are limited in data modalities they can use. Here, we developed a deep learning framework (EIR) for PRS prediction which includes a model, genome-local-net (GLN), specifically designed for large scale genomics data. The framework supports multi-task (MT) learning, automatic integration of other clinical and biochemical data, and model explainability. When applied to individual level data in the UK Biobank, we found that GLN outperformed LASSO for a wide range of diseases and in particularly autoimmune diseases. Furthermore, we show that this was likely due to modelling epistasis, and we showcase this by identifying widespread epistasis for Type 1 Diabetes. Furthermore, we trained PRS by integrating genotype, blood, urine and anthropometrics and found that this improved performance for 93% of 290 diseases and disorders considered. Finally, we found that including genotype data provided better calibrated PRS models compared to using measurements alone. EIR is available at https://github.com/arnor-sigurdsson/EIR.

bioinformatics↗

bioRxiv2017

Text mining of 15 million full-text scientific articles

Across academia and industry, text mining has become a popular strategy for keeping up with the rapid growth of the scientific literature. Text mining of the scientific literature has mostly been carried out on collections of abstracts, due to their availability. Here we present an analysis of 15 million English scientific full-text articles published during the period 1823-2016. We describe the development in article length and publication sub-topics during these nearly 250 years. We showcase the potential of text mining by extracting published protein-protein, disease-gene, and protein subcellular associations using a named entity recognition system, and quantitatively report on their accuracy using gold standard benchmark data sets. We subsequently compare the findings to corresponding results obtained on 16.5 million abstracts included in MEDLINE and show that text mining of full-text articles consistently outperforms using abstracts only.

bioinformatics↗