Skip to content
Log in
Sponsored Have an interesting project or opportunity? → Hire me ↗
Sponsored Design Multi-Agent AI Systems Using MCP and A2A: Engineer your own Pyt Design Multi-Agent AI Systems Using MCP and A2A: Engineer your own Pyt

Author

L. Milani

View on Semantic Scholar

2 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Open access Jul 2026

Psychiatric genetic liability is associated with the severity of COVID-19 and other acute respiratory infections: an observational study across five Northern European countries

Pre-existing psychiatric disorders have been associated with the severity of acute respiratory infections, including COVID-19, particularly in hospitalized populations. However, the underlying mechanisms, especially in community-based populations, remain unknown, limiting preparedness for future pandemics. We investigated the role of genetic liability for psychiatric disorders and related traits in COVID-19 and other respiratory infection severity among individuals reporting SARS-CoV-2 testing and available respiratory symptom data. We included population-based cohort data from Denmark, Estonia, Iceland, Norway, and the United Kingdom (N = 78,507; 62% female; 41% SARS-CoV-2-positive; May 2020—March 2023). Respiratory infection severity was defined by self-reported days bedridden and an eleven-symptom composite score during the acute illness phase. Polygenic scores (PGSs) for eight psychiatric disorders and the personality trait neuroticism indexed genetic liability. Meta-analysed estimates from stepwise-adjusted regression analyses controlling for education, lifestyle, and psychiatric and somatic diagnosis history were used to assess associations between PGSs and acute respiratory infection severity, stratified by COVID-19 status. Higher genetic liability to insomnia, major depression (MD), and neuroticism was consistently associated with longer time bedridden (relative risk ratios per 1 SD change in PGS: 1.07–1.14, 95% confidence intervals [CIs]: 1.03–1.21) and more symptoms (incident rate ratios [IRRs]: 1.02–1.04, 95% CIs: 1.01–1.05) due to COVID-19. PGSs for insomnia, MD, neuroticism, anxiety disorder, bipolar disorder, and schizophrenia were also linked to higher non-COVID-19 symptom count (IRRs: 1.01–1.05, 95% CIs: 1.00–1.07). These associations remained robust after adjustment for lifestyle, education and medical comorbidities. Individuals in the highest MD PGS decile reported 20% more symptoms in non-COVID-19 illness and 10% more in COVID-19 compared with the bottom decile. No consistent associations were detected with attention-deficit/hyperactivity disorder, alcohol dependence and post-traumatic stress disorder. No genetic link was observed with ischaemic heart disease or type 2 diabetes. Shared genetics may contribute to the link between psychiatric conditions and respiratory infection severity in community settings. Our results underscore the role of psychiatric genetic liability, beyond diagnosed psychiatric disorders, in contributing to both COVID-19 and other respiratory infection severity. These findings provide insights that may improve future risk stratification and public health strategies targeting respiratory viruses.

K. Kõiv, R. Askeland, L. N. Christoffersen et al. · 0 citations
Open access Aug 2026

Detecting CYP2C19 deletions from genotyping array signals using neural networks

Since copy number variations (CNVs) in pharmacogenes can cause significant alterations in drug metabolism, their reliable detection is of high importance both for large-scale studies and personalized medicine. Whole-genome sequencing, and specifically long-read sequencing, is the gold standard for CNV detection. Despite increasing availability of these technologies, genotyping arrays are still widely used as cost-effective alternatives in biobank and clinical settings, yet calling CNVs based on array intensity signals is challenging due to low base pair resolution. In this work, we developed a neural network model, nnCNV, to predict deletions in the CYP2C19 pharmacogene region from array intensity signals. We compared our method to the most widely used algorithm, PennCNV, and demonstrated better performance reaching 100% accuracy in the test dataset. Furthermore, we predicted probe-by-probe CYP2C19 deletion coordinates for all Estonian Biobank samples using nnCNV and PennCNV, and validated these predictions using an identity-by-descent (IBD) sharing method, which also demonstrated superior nnCNV performance. For the deletion samples with conflicting PennCNV and nnCNV predictions, we performed PCR analysis for validation, which showed 97% precision for nnCNV compared to 23% for PennCNV. Finally, we assessed the gradient-based feature importance maps and showed that nnCNV utilizes signal intensity information not only from deletion probes, but also from probes in flanking regions. Our results demonstrate that long-range information, which cannot be utilized by hidden Markov models, can improve CNV calling.

Burak Yelmen, R. Hofmeister, Viido Kaur Lutsar et al. · 0 citations