The EN-TEx resource of multi-tissue personal epigenomes & variant-impact models
Joel Rozowsky, Jiahao Gao, Beatrice Borsari, Yucheng Yang, Timur R. Galeev, Gamze Gürsoy, Charles B. Epstein, Kun Xiong, Jinrui Xu, Tianxiao Li, Jason Liu, Keyang Yu, Ana Berthel, Zhanlin Chen, Fábio C. P. Navarro, Maxwell S. Sun, James C. Wright, Justin Chang, Christopher J. F. Cameron, Noam Shoresh, Elizabeth Gaskell, Jörg Drenkow, Jessika Adrian, Sergey Aganezov, François Aguet, Gabriela Balderrama-Gutierrez, Samridhi Banskota, Guillermo Barreto Corona, Sora Chee, Surya B. Chhetri, Gabriel Conte Cortez Martins, Cassidy Danyko, Carrie Davis, Daniel Farid, Nina Farrell, Idan Gabdank, Yoel Gofin, David U. Gorkin, Mengting Gu, Vivian Hecht, Benjamin C. Hitz, Robbyn Issner, Yunzhe Jiang, Melanie Kirsche, Xiangmeng Kong, Bonita R. Lam, Shantao Li, Bian Li, Xiqi Li, Khine Lin, Ruibang Luo, Mark Mackiewicz, Ran Meng, Jill E. Moore, Jonathan M. Mudge, Nicholas Nelson, Chad Nusbaum, Ioann Popov, Henry Pratt, Yunjiang Qiu, Srividya Ramakrishnan, Joe Raymond, Leonidas Salichos, Alexandra Scavelli, Jacob Schreiber, Fritz J. Sedlazeck, Lei Hoon See, Rachel M. Sherman, Xu Shi, Minyi Shi, Cricket A. Sloan, J. Seth Strattan, Zhen Tan, Forrest Y. Tanaka, Anna Vlasova, Jun Wang, Jonathan M. Werner, Brian A. Williams, Min Xu, Chengfei Yan, Yu Lu, Christopher Zaleski, Jing Zhang, Kristin Ardlie, J. Michael Cherry, Eric M. Mendenhall, William Stafford Noble, Zhiping Weng, Morgan E. Levine, Alexander Dobin, B Wold, A Mortazavi, Bing Ren, Jesse Gillis, R Myers, M Snyder, Jyoti S. Choudhary, Aleksandar Milosavljevic, Michael C. Schatz, B Bernstein
Abstract
Understanding how genetic variants impact molecular phenotypes is a key goal of functional genomics, currently hindered by reliance on a single haploid reference genome. Here, we present the EN-TEx resource of 1,635 open-access datasets from four donors (∼30 tissues × ∼15 assays). The datasets are mapped to matched, diploid genomes with long-read phasing and structural variants, instantiating a catalog of >1 million allele-specific loci. These loci exhibit coordinated activity along haplotypes and are less conserved than corresponding, non-allele-specific ones. Surprisingly, a deep-learning transformer model can predict the allele-specific activity based only on local nucleotide-sequence context, highlighting the importance of transcription-factor-binding motifs particularly sensitive to variants. Furthermore, combining EN-TEx with existing genome annotations reveals strong associations between allele-specific and GWAS loci. It also enables models for transferring known eQTLs to difficult-to-profile tissues (e.g., from skin to heart). Overall, EN-TEx provides rich data and generalizable models for more accurate personal functional genomics.