From eathomp at uw.edu Wed Nov 16 20:51:16 2022 From: eathomp at uw.edu (Elizabeth Thompson) Date: Wed Mar 20 12:11:55 2024 Subject: [PopGenLunch] Fwd: [Statdept] Aaron Baraff - Final Exam - Monday, November 21, 2022 @ 3pm In-Reply-To: References: Message-ID: Hi All: Please see below re Aaron Baraff's Final Exam presentation next Monday 3:30 pm in C301 in Padelford. While the presentation will be hybrid to allow for all contingencies,we are hoping for at least some to come in person to enjoy this talk -- a talk very relevant to probabilistic modeling and inference in both StatGen and PopGen.. Thanks, Elizabeth ---------- Forwarded message --------- *FINAL EXAM* *Aaron Baraff* *Likelihood-based haplotype frequency modeling using variable-order Markov chains* [image: Gold Boundless Bar] *Date: Monday, November 21, 2022 Time: 3:00 p.m. Location: Padelford Hall, Room C-301 Zoom Link: https://washington.zoom.us/j/4206552403 [washington.zoom.us] * * Advisor: Elizabeth Thompson* [image: Gold Boundless Bar] *Abstract:* The localized haplotype-cluster model uses variable-order Markov chains (VOMCs) to create an empirical model for haplotype probabilities that adapts to the changing structure of linkage disequilibrium (LD) across the genome. By clustering partial haplotypes based on the Markov property as represented by a directed acyclic graph (DAG), the model is able to take advantage of context-sensitive conditional independencies to improve estimates of haplotype frequencies while still respecting the dependencies induced by LD. We introduce a method for training such models using regularized likelihood functions to prevent overfitting along with a method for cross-validation to select a regularization parameter which accounts for the high probability of out-of-sample haplotypes not accommodated by the model. When applied to dense single nucleotide polymorphism (SNP) markers from population data, our method obtains a better-fitting and more parsimonious model than the leading method. In addition, we note that these models represent a VOMC defined in a single direction along the genome, which ignores the LD structure that could be represented by conditional independencies in the opposite direction. Therefore, fitting the model to the same data in the reverse direction along the genome usually results in different haplotype frequency estimates, which is an undesirable property for genomic models. We develop a method of reconciling two DAG models fit in opposite directions along the genome that takes advantage of the differing LD structure represented in both models to derive a new bidirectional model. When trying to detect segments of identity by descent (IBD) among individuals, background LD can be a source of noise that obfuscates haplotypic similarity due to recent coancestry. Methods of IBD segment detection that do not account for LD can have a high false positive rate. We introduce a method for IBD segment detection using a hidden Markov model (HMM) that incorporates a DAG model in the hidden layer to adjust for LD. Unlike similar methods, ours models the full set of 15 IBD states among the four chromosomes of two individuals. When applied to simulated dense SNP marker data, our method provides more accurate IBD segment detection than other leading methods. ?The University of Washington is committed to providing access, equal opportunity and reasonable accommodation in its services, programs, activities, education and employment for individuals with disabilities. To request disability accommodation contact the Disability Services Office at least ten days in advance at: 206-543-6450/V, 206-543-6452/TTY, 206-685-7264 (FAX), or dso@u.washington.edu.? _______________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 255 bytes Desc: not available URL: From eathomp at uw.edu Wed Nov 16 20:53:25 2022 From: eathomp at uw.edu (Elizabeth Thompson) Date: Wed Mar 20 12:11:55 2024 Subject: [PopGenLunch] [Statdept] Aaron Baraff - Final Exam - Monday, November 21, 2022 @ 3pm In-Reply-To: References: Message-ID: Oops. --- 3:00 pm !!!! On Wed, Nov 16, 2022 at 8:51 PM Elizabeth Thompson wrote: > > Hi All: > > Please see below re Aaron Baraff's Final Exam presentation next Monday > 3:30 pm in C301 in Padelford. While the presentation will be hybrid to > allow for all contingencies,we are hoping for at least some to come in > person to enjoy this talk -- a talk very relevant to probabilistic modeling > and inference in both StatGen and PopGen.. > > Thanks, > Elizabeth > > > ---------- Forwarded message --------- > > > *FINAL EXAM* > > *Aaron Baraff* > > > > *Likelihood-based haplotype frequency modeling using variable-order Markov > chains* > > [image: Gold Boundless Bar] > > > > > *Date: Monday, November 21, 2022 Time: 3:00 p.m. Location: Padelford Hall, > Room C-301 Zoom Link: https://washington.zoom.us/j/4206552403 > [washington.zoom.us] > * > * Advisor: Elizabeth Thompson* > > [image: Gold Boundless Bar] > > *Abstract:* > > The localized haplotype-cluster model uses variable-order Markov chains > (VOMCs) to create an empirical model for haplotype probabilities that > adapts to the changing structure of linkage disequilibrium (LD) across the > genome. By clustering partial haplotypes based on the Markov property as > represented by a directed acyclic graph (DAG), the model is able to take > advantage of context-sensitive conditional independencies to improve > estimates of haplotype frequencies while still respecting the dependencies > induced by LD. We introduce a method for training such models using > regularized likelihood functions to prevent overfitting along with a method > for cross-validation to select a regularization parameter which accounts > for the high probability of out-of-sample haplotypes not accommodated by > the model. When applied to dense single nucleotide polymorphism (SNP) > markers from population data, our method obtains a better-fitting and more > parsimonious model than the leading method. > > > > In addition, we note that these models represent a VOMC defined in a > single direction along the genome, which ignores the LD structure that > could be represented by conditional independencies in the opposite > direction. Therefore, fitting the model to the same data in the reverse > direction along the genome usually results in different haplotype frequency > estimates, which is an undesirable property for genomic models. We develop > a method of reconciling two DAG models fit in opposite directions along the > genome that takes advantage of the differing LD structure represented in > both models to derive a new bidirectional model. > > > > When trying to detect segments of identity by descent (IBD) among > individuals, background LD can be a source of noise that obfuscates > haplotypic similarity due to recent coancestry. Methods of IBD segment > detection that do not account for LD can have a high false positive rate. > We introduce a method for IBD segment detection using a hidden Markov model > (HMM) that incorporates a DAG model in the hidden layer to adjust for LD. > Unlike similar methods, ours models the full set of 15 IBD states among the > four chromosomes of two individuals. When applied to simulated dense SNP > marker data, our method provides more accurate IBD segment detection than > other leading methods. > > > > > > ?The University of Washington is committed to providing access, equal > opportunity and reasonable accommodation in its services, programs, > activities, education and employment for individuals with disabilities. To > request disability accommodation contact the Disability Services Office at > least ten days in advance at: 206-543-6450/V, 206-543-6452/TTY, > 206-685-7264 (FAX), or dso@u.washington.edu.? > > > _______________________________________________ > > -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 255 bytes Desc: not available URL: From eathomp at uw.edu Thu Nov 17 22:40:29 2022 From: eathomp at uw.edu (Elizabeth Thompson) Date: Wed Mar 20 12:11:55 2024 Subject: [PopGenLunch] PIMS course in Mathematical Population Genetics and Genomics Message-ID: Aillene MacPherson (https://www.sfu.ca/math/people/faculty/ailenem.html), at SFU is giving a PIMS course (https://courses.pims.math.ca/courses/2022-2023/mathematical-population-genetics-and-genomics/) on the mathematics of evolution next semester. Since UW is a member of PIMS (https://www.pims.math.ca/), Ailene will welcome UW students. Admin is a bit complicated, and students would register for UW Independent study/Special topics credit, which I have volunteered to coordinate, but this is a terrific opportunity for PhD students in this area. If you are interested, or have students who should be interested, please contact me, and we can discover how to work this. I attach a course flyer, but for UW students, contact me, rather than Ailene, until we know we can do this. Best, Elizabeth -------------- next part -------------- A non-text attachment was scrubbed... Name: APMA990_Syllabus.pdf Type: application/pdf Size: 229811 bytes Desc: not available URL: