ICCFGG 2022
#4 Ultracontinuous Genomes Elucidate Complex Speciation Patterns within Panthera Andrew J. Harris 1,2 , Brian W. Davis 1 , 2 , Klaus-Peter Koepfli 3,4 , Eduardo Eizirik 5 , William J. Murphy 1,2 ajharris@cvm.tamu.edu 1 Veterinary Integrative Biosciences, Texas A&M University, College Station, TX, USA, 2 Interdisciplinary Program in Genetics & Genomics, Texas A&M University, College Station, TX,USA, 3 Smithsonian Con- servation Biology Institute, Center for Species Survival, National Zoological Park, Washington, DC, USA, 4 Smithsonian-Mason School of Conservation, George Mason University, Front Royal, VA, USA, 5 School of Health and Life Sciences, Pontifical Catholic University of Rio Grande do Sul, Rio Grande do Sul, Brazil Genomes are a mosaic of evolutionary histories that reflect ancient signatures of true species relationships and incomplete lineage sorting (ILS) or gene flow. Understanding how and why phylogenetic signal varies across species genomes can yield powerful insights into species’ evolutionary histories and adaptive evolution. By integrating diverse data types with local genealogies, one can differentiate genetic variation consistent with the species tree from that stemming from natural selection, ILS, or gene flow. To better resolve the complex evolutionary history of the living cat species of the genus Panthera, we aligned PacBio® HiFi genomes from the jaguar, leopard, snow leopard, lion, and Indochinese clouded leopard to a highly continuous single haplotype assembly from the tiger. We conducted a whole-genome sliding window phylogenomic analysis and used our novel phylogenomics browser Tree House Explorer to visualize genome-wide variation in evolutionary histories and genetic divergence on a chromosome-by-chromosome basis in the context of recombination rates. We identified significant differences in the distribution of phylogenetic signal along the X chromosome, where low recombining regions harbor signal of the species tree. However, a subset of these regions contain a signal of ancient introgression of the leopard with an extinct or unsampled species with an ancient origin. Whole genome evaluations of structural variation also indicate an enrichment of structural rearrangements along the X chro- mosome, which likely played a role in the unique distribution of phylogenetic signal across the X chromosome and reproductive isolation. #5 Deep learning approach to predict the impact of canine regulatory mutations Christophe Hitte 1 ,Camille Kergal 1 , Doga2 Consortium, Marie-Dominique Galibert 1 , Thomas Derrien 1 hitte@univ-rennes1.fr 1 Univ Rennes 1, CNRS, IGDR – UMR6290, F-35000 Rennes, France, 2 DogA: www.doggenomeannotation.org Deep neural networks have been recently shown to be powerful methods to predict gene expression and, in fine, to assess the impact of regulatory mutations on gene expression. Here, we used the human-based tool Basenji to train a dog-specific model of gene expression based on deep-learn- ing (DL) to predict the impact of non-coding mutations on gene expression dedicated to the dog genome. We trained the model with a comprehensive set of canine Cap Analysis of Gene Expression data representative of 37 core canine tissues. We showed that the dog-specific model reached
30
Made with FlippingBook - Online magazine maker