<html><body><div id="zimbraEditorContainer" style="font-family: garamond,new york,times,serif; font-size: 12pt; color: #000000" class="2"><div></div><div data-marker="__QUOTED_TEXT__"><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">
Max Homilius will present his FPO, "Network-Based Prioritization of Disease Genes, Animal Models, and Drug Targets" on Friday, 8/24/2018 at 10am, 280 Lewis-Sigler Institute.</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">The members of his committee are as follows: Readers: Barbara Engelhardt and Tilo Grosser (UPenn); Examiners: Mona Singh, Coleen Murphy (MOL), and Olga Troyanskaya (adviser).<br></p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">A copy of his thesis is available upon request.</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">Everyone is invited to attend his talk. The talk abstract follows below.</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;"><br data-mce-bogus="1"></p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">In living organisms, biomolecules interact in complex molecular networks that underlie</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">cell function and whose dysregulation leads to disease. These networks thus</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">provide a key lens for understanding the molecular basis of human disease as well</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">as treatment development. In this thesis, I develop three network-based computational</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">approaches for human disease research and gaining insight into the mode of</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">action of drugs. At their core, these methods employ functional interaction networks,</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">which provide a genome-wide view of biochemical and pathway-level interactions and</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">summarize essential functional information derived from diverse and heterogeneous</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">functional genomics experiments.</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">First, I propose a network-based method that can detect critical genes and pathways</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">targeted by a drug treatment from gene expression data even in the absence</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">of large-scale expression diāµerences. This approach enables the analysis of low-dose</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">drug screens, ranking potential targets and drug-perturbed biological processes with</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">higher accuracy than prior network-based methods or gene-expression data alone.</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">Furthermore, I present a method that by inferring and comparing genome-wide profiles</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">for human diseases and animal model phenotypes identifies analogous disease</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">models with high accuracy and more robustly than prior methods relying on shared</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">gene content. This method allows to aggregate the wealth of existing model organism</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">knowledge across multiple species and to identify related phenotypes and novel homologous</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">genes of human diseases for experimental follow-up. Lastly, by constructing</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">a joint tissue-specific classifier for human disease genes, we can significantly improve</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">the prediction of associated genes for rare human diseases. This neural network-based</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">approach makes use of a functional network embedding leveraging tissue-specific expression</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">data and model organism phenotype information in a multi-label classification</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">setting. Overall, the methods I developed provide data-driven, molecular-level</p><p style="margin-left: 50px;" data-mce-style="margin-left: 50px;">solutions to major biological challenges relevant to human health.</p></div></div></body></html>