Timezone: »

 
Sample-Specific Contextualized Graphical Models Using Clinical and Molecular Data Reveal Transcriptional Network Heterogeneity Across 7000 Tumors
Caleb Ellington · Ben Lengerich · Thomas Watkins · Jiekun Yang · Manolis Kellis · Eric Xing

Fri Dec 09 03:48 PM -- 04:00 PM (PST) @
in Graph Learning for Industrial Applications: Finance, Crime Detection, Medicine and Social Media »

Cancers are shaped by somatic mutations, microenvironment, and patient background, each altering both gene expression and gene regulatory networks (GRNs) in complex ways, resulting in highly-variable cellular states and dynamics. Inferring GRNs from expression data can help characterize this regulation-driven heterogeneity, but network inference is intractable without many statistical samples, limiting GRNs to cluster-level analyses that ignore intra-cluster heterogeneity. We propose to move beyond cluster-based analyses by using \emph{contextualized} learning, a meta-learning paradigm, to generate sample-specific network models from sample contexts. We unify three network classes (correlation, Markov, Bayesian) and estimate sample-specific GRNs for 7000 tumours across 25 tumor types, with each network contextualized by copy number and driver mutation profiles, tumor microenvironment and patient demographics. Sample-specific networks provide a de-noised view of gene expression dynamics at sample-specific resolution, which reveal co-expression modules in correlation networks (CNs), clique structures and neighborhood selection in Markov Networks (MNs), and causal ordering and probability factorization in Bayesian Networks (BNs). Sample-specific networks enable GRN-based precision oncology, including brain tumor subtyping that improves survival prognosis.

[ Poster] 

Author Information

Caleb Ellington (Carnegie Mellon University)
Ben Lengerich (MIT)
Thomas Watkins (MIT)
Jiekun Yang (MIT)
Manolis Kellis (Massachusetts Institute of Technology)
Manolis Kellis

Manolis Kellis is a professor of computer science and artificial intelligence at MIT, and leads the MIT Computational Biology Group at MIT CSAIL and the Broad Institute of MIT and Harvard. His research seeks to understand the mechanistic basis of human disease, to develop new therapeutics that reverse dis-ease circuitry, and to enable personalized medicine, using AI and machine learning to integrate genetics and genomics, single-cell epigenomics and transcriptomics, and high-throughput experiments, applied to Alzheimer's, Obesity, Cancer, Schizophrenia, Cardiovascular, and Immune Disorders. He helped lead several large-scale genomics projects, including Roadmap Epigenomics, ENCODE, Genotype Tissue-Expression (GTEx), and Comparative Genomics projects. He has authored over 280 journal publications cited more than 160,000 times. He received the US Presidential Award for Science and Engineering by Barack Obama, the Mendel Medal for Outstanding Achievements in Science, the NIH Director’s Transformative Research Award, the Argo Science Award by the Hellenic President, the Boston Patent Law Association award, the NSF CAREER award, the Alfred P. Sloan Fellowship, the Technology Review TR35 recognition, the AIT Niki Award, and the Sprowls award for the best Ph.D. thesis in computer science at MIT. He has obtained more than 20 multi-year grants from the NIH, and his trainees hold faculty positions at Stanford, Harvard, CMU, McGill, Johns Hopkins, UCLA, and other top universities. He lived in Greece and France before moving to the US, and he studied and conducted research at MIT, the Xerox Palo Alto Research Center, and the Cold Spring Harbor Lab. For more info, see: compbio.mit.edu

Eric Xing (Petuum Inc. / Carnegie Mellon University)

More from the Same Authors