BEGIN:VCALENDAR VERSION:2.0 PRODID:-//wp-events-plugin.com//7.2.3.1//EN TZID:America/Los_Angeles X-WR-TIMEZONE:America/Los_Angeles BEGIN:VEVENT UID:326@escience.washington.edu DTSTART;TZID=America/Los_Angeles:20260305T163000 DTEND;TZID=America/Los_Angeles:20260305T172000 DTSTAMP:20260311T171226Z URL:https://escience.washington.edu/events/uw-data-science-seminar-steven- roberts-and-kathleen-durkin/ SUMMARY:UW Data Science Seminar: Steven Roberts and Kathleen Durkin DESCRIPTION:Please join us for a UW Data Science Seminar featuring UW Aquat ic and Fishery Sciences Professor Steven Roberts and Graduate Student Kath leen Durkin on Thursday\, March 5th from 4:30 to 5:20 p.m. PT. The seminar will be held in IEB G109.\n\n \;\n"Sparse Tensor Decomposition and El astic Net Regression for Multi-Species Gene Expression Analysis"\nAbstract : Classical transcriptomic methods -- pairwise differential expression (DE Seq2/edgeR)\, PCA\, and WGCNA -- are fundamentally two-dimensional and can not natively represent datasets structured along three simultaneous axes: genes\, samples\, and time. When applied to multi-species time-series data \, these approaches require flattening or ignoring at least one axis\, lea ding to information loss\, arbitrary thresholding\, and post-hoc compariso ns that lack statistical coherence.\n\nHere we apply sparse CP tensor deco mposition\, implemented in the Barnacle framework\, to E5 timeseries coral transcriptomic data spanning three evolutionarily divergent coral species (Acropora pulchra\, Porites evermanni\, and Porites tuahiniensis) acros s four timepoints (TP1--TP4). Gene expression was quantified at the orthol og-group level\, constructing a three-mode tensor (ortholog groups × spec ies-samples × timepoints) and normalized using sctransform. Details of th e E5-MOSAiC project can be found here.\n\nBuilding on these tensor decompo sition results\, we further applied an Elastic Net (EN) regression pipelin e to predict the expression using multi-omic epigenetic predictors. Gene s ets identified from individual tensor components were then used as respons e variables in Elastic Net regression models\, with three classes of epige netic predictors. Our results identify specific miRNAs\, lncRNAs\, and CpG sites that are the strongest predictors of gene expression variation with in each species\, providing evidence for multi-layered epigenetic regulati on of transcription in corals across seasonal time scales.\n\nThis work de monstrates that sparse tensor decomposition is a principled\, fully unsupe rvised approach for identifying conserved and species-specific gene co-exp ression programs across multi-species\, multi-timepoint molecular datasets \, a class of problem increasingly common in environmental and comparative genomics\n\n \;\n\nSpeaker Bios: Steven Roberts is a Professor in the School of Aquatic and Fishery Sciences at the University of Washington\, where his research advances environmental memory as a foundational princip le linking molecular regulation\, physiology\, and resilience in marine or ganisms. His work integrates genomics\, epigenetics\, and organismal physi ology to understand how past environmental conditions shape future perform ance across individuals\, populations\, and generations. Through a focus o n marine invertebrates\, his research bridges discovery and application\, informing conservation\, restoration\, and sustainable aquaculture in chan ging coastal ecosystems.\n\nKathleen Durkin is a graduate student in Aquat ic and Fishery Sciences with broad interests in environmental responses in marine invertebrates. As an undergraduate at Harvey Mudd College\, she st udied soft coral taxonomy and population genetics using single- and multi- locus sequencing approaches\, graduating in 2023 with a B.S. in Mathematic al and Computational Biology. Her current research includes studying non-c oding RNAs in corals and multigenerational DNA methylation patterns in Eas tern oysters. Kathleen maintains a publicly accessible digital lab noteboo k that documents her ongoing research and reflects a commitment to open sc ience practices.\n\n\nThe 2025-2026 seminars will be held in person\, and are free and open to the public.\n\n ATTACH;FMTTYPE=image/jpeg:https://escience.washington.edu/wp-content/uploa ds/2026/02/Untitled-design-1.jpg END:VEVENT BEGIN:VTIMEZONE TZID:America/Los_Angeles X-LIC-LOCATION:America/Los_Angeles BEGIN:STANDARD DTSTART:20251102T010000 TZOFFSETFROM:-0700 TZOFFSETTO:-0800 TZNAME:PST END:STANDARD END:VTIMEZONE END:VCALENDAR