BEGIN:VCALENDAR VERSION:2.0 PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4// BEGIN:VEVENT UID:20260224T071418EST-1432o3sses@132.216.98.100 DTSTAMP:20260224T121418Z DESCRIPTION:Kata Vuk\, PhD\n\nFaculty of Informatics and Data Science | Uni versity of Regensburg\n\nVisiting Scholar | 91ºÚÁÏÍø \n\n \n\nWHE N: Wednesday\, March 4\, 2026\, from 3:30 to 4:30 p.m.\n WHERE: Hybrid | 20 01 91ºÚÁÏÍø College Avenue\, Rm 1140\; Zoom\n NOTE: Kata Vuk will be presenti ng in-person at SPGH \n\nAbstract\n\nGenome-wide association studies (GWAS ) identify trait-associated loci\, but translating these signals into caus al genes remains statistically challenging. I present a stratified modelin g framework for post-GWAS gene prioritization that accounts for heterogene ity in variant-to-gene mechanisms. Genes are classified as regulatory\, in tragenic\, or mixed\, and setting-aware features are integrated within a s ingle random forest model. This design enables the model to learn mechanis m-specific patterns while maintaining comparability of predicted probabili ties across genes\, and it exposes substantial differences in predictive a ccuracy across gene settings. Applied to kidney function (eGFR)\, the fram ework uncovers systematic differences in the genomic features driving prio ritization. Locus-level analyses demonstrate stable top-gene identificatio n across bootstrap resampling and model comparisons.\n\nSpeaker Bio\n\nKat a Vuk is currently visiting 91ºÚÁÏÍø as an IVADO Visiting Scholar . Since January 2025\, she has been supported by a fellowship from the Bav arian Research Institute for Digital Transformation in Germany. She is a p ostdoctoral researcher at the Chair of Machine Learning at the University of Regensburg and received her PhD in Mathematics from Ruhr University Boc hum. Her research lies at the intersection of statistics and machine learn ing\, with a focus on interpretable and personalized modeling\, change-poi nt analysis\, and their applications in biostatistics\, genetics\, and bio medical research. \n DTSTART:20260304T203000Z DTEND:20260304T213000Z SUMMARY:Stratified Modeling for Causal Gene Prioritization with Interpretab le Tree Ensembles URL:/spgh/channels/event/stratified-modeling-causal-ge ne-prioritization-interpretable-tree-ensembles-371251 END:VEVENT END:VCALENDAR