Faculty of Science, Technology and Medicine
Faculty of Law, Economics and Finance
Faculty of Humanities, Education and Social Sciences
Interdisciplinary Centre for Security, Reliability and Trust
Luxembourg Centre for Systems Biomedicine
Luxembourg Centre for Contemporary and Digital History
Luxembourg Centre for European Law
Luxembourg Centre for Socio-Environmental Systems
This is a hybrid event. Join in person at the LCSB, or tune in via Webex here.
Genome-wide variant effect prediction
Understanding the functional impact of genetic variation is a central challenge in human genetics. With millions of variants identified in each human genome, distinguishing pathogenic mutations from benign variation is an important prerequisite for clinical interpretation and personalized medicine. This lecture will present computational approaches for genome-wide variant effect prediction, focusing on the Combined Annotation Dependent Depletion (CADD) framework. CADD integrates diverse genomic annotations, including conservation metrics, functional genomics data, and protein structure predictions, into unified deleteriousness scores applicable across coding and non-coding regions. The presentation will discuss the machine learning strategies underlying these predictions, the use of evolutionary signatures as training objectives, and recent advances incorporating deep learning-derived splice predictions and protein language models. The talk will also address practical applications in clinical variant interpretation and ongoing efforts to improve prediction accuracy across diverse variant types and molecular mechanisms.
About the speaker
Martin Kircher is Professor of Regulatory Genomics at the University of Lübeck and University Medical Center Schleswig-Holstein, and a Fellow at the Berlin Institute of Health at Charité, where he heads the Computational Genome Biology research group. His research focuses on computational methods for identifying functionally relevant genomic sequences and predicting the effects of genetic variants. He developed and maintains CADD (Combined Annotation Dependent Depletion), one of the most widely used tools for genome-wide variant effect prediction. His group’s work spans variant scoring tools for different molecular mechanisms, analysis of massively parallel reporter assays, and computational methods for cell-free DNA analysis.
The Causal Analysis of Biomedical Data Lecture Series is supported by the Luxembourg National Research Fund (FNR) RESCOM Program.
