Yifei Chen1, Yi Li2, Rajiv Narayan3, Aravind Subramanian3, Xiaohui Xie4. 1. Department of Computer Science, University of California, Irvine, CA 92697, USA Baidu Research-Big Data Lab, Beijing, 100085, China. 2. Department of Computer Science, University of California, Irvine, CA 92697, USA. 3. Broad Institute of MIT And Harvard, Cambridge, MA 02142, USA. 4. Department of Computer Science, University of California, Irvine, CA 92697, USA Center for Complex Biological Systems, University of California, Irvine, CA 92697, USA.
Abstract
MOTIVATION: Large-scale gene expression profiling has been widely used to characterize cellular states in response to various disease conditions, genetic perturbations, etc. Although the cost of whole-genome expression profiles has been dropping steadily, generating a compendium of expression profiling over thousands of samples is still very expensive. Recognizing that gene expressions are often highly correlated, researchers from the NIH LINCS program have developed a cost-effective strategy of profiling only ∼1000 carefully selected landmark genes and relying on computational methods to infer the expression of remaining target genes. However, the computational approach adopted by the LINCS program is currently based on linear regression (LR), limiting its accuracy since it does not capture complex nonlinear relationship between expressions of genes. RESULTS: We present a deep learning method (abbreviated as D-GEX) to infer the expression of target genes from the expression of landmark genes. We used the microarray-based Gene Expression Omnibus dataset, consisting of 111K expression profiles, to train our model and compare its performance to those from other methods. In terms of mean absolute error averaged across all genes, deep learning significantly outperforms LR with 15.33% relative improvement. A gene-wise comparative analysis shows that deep learning achieves lower error than LR in 99.97% of the target genes. We also tested the performance of our learned model on an independent RNA-Seq-based GTEx dataset, which consists of 2921 expression profiles. Deep learning still outperforms LR with 6.57% relative improvement, and achieves lower error in 81.31% of the target genes. AVAILABILITY AND IMPLEMENTATION: D-GEX is available at https://github.com/uci-cbcl/D-GEX CONTACT: xhx@ics.uci.edu SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.
MOTIVATION: Large-scale gene expression profiling has been widely used to characterize cellular states in response to various disease conditions, genetic perturbations, etc. Although the cost of whole-genome expression profiles has been dropping steadily, generating a compendium of expression profiling over thousands of samples is still very expensive. Recognizing that gene expressions are often highly correlated, researchers from the NIH LINCS program have developed a cost-effective strategy of profiling only ∼1000 carefully selected landmark genes and relying on computational methods to infer the expression of remaining target genes. However, the computational approach adopted by the LINCS program is currently based on linear regression (LR), limiting its accuracy since it does not capture complex nonlinear relationship between expressions of genes. RESULTS: We present a deep learning method (abbreviated as D-GEX) to infer the expression of target genes from the expression of landmark genes. We used the microarray-based Gene Expression Omnibus dataset, consisting of 111K expression profiles, to train our model and compare its performance to those from other methods. In terms of mean absolute error averaged across all genes, deep learning significantly outperforms LR with 15.33% relative improvement. A gene-wise comparative analysis shows that deep learning achieves lower error than LR in 99.97% of the target genes. We also tested the performance of our learned model on an independent RNA-Seq-based GTEx dataset, which consists of 2921 expression profiles. Deep learning still outperforms LR with 6.57% relative improvement, and achieves lower error in 81.31% of the target genes. AVAILABILITY AND IMPLEMENTATION: D-GEX is available at https://github.com/uci-cbcl/D-GEX CONTACT: xhx@ics.uci.edu SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.
Authors: Charles Wang; Binsheng Gong; Pierre R Bushel; Jean Thierry-Mieg; Danielle Thierry-Mieg; Joshua Xu; Hong Fang; Huixiao Hong; Jie Shen; Zhenqiang Su; Joe Meehan; Xiaojin Li; Lu Yang; Haiqing Li; Paweł P Łabaj; David P Kreil; Dalila Megherbi; Stan Gaj; Florian Caiment; Joost van Delft; Jos Kleinjans; Andreas Scherer; Viswanath Devanarayan; Jian Wang; Yong Yang; Hui-Rong Qian; Lee J Lancashire; Marina Bessarabova; Yuri Nikolsky; Cesare Furlanello; Marco Chierici; Davide Albanese; Giuseppe Jurman; Samantha Riccadonna; Michele Filosi; Roberto Visintainer; Ke K Zhang; Jianying Li; Jui-Hua Hsieh; Daniel L Svoboda; James C Fuscoe; Youping Deng; Leming Shi; Richard S Paules; Scott S Auerbach; Weida Tong Journal: Nat Biotechnol Date: 2014-08-24 Impact factor: 54.908
Authors: Tuuli Lappalainen; Michael Sammeth; Marc R Friedländer; Peter A C 't Hoen; Jean Monlong; Manuel A Rivas; Mar Gonzàlez-Porta; Natalja Kurbatova; Thasso Griebel; Pedro G Ferreira; Matthias Barann; Thomas Wieland; Liliana Greger; Maarten van Iterson; Jonas Almlöf; Paolo Ribeca; Irina Pulyakhina; Daniela Esser; Thomas Giger; Andrew Tikhonov; Marc Sultan; Gabrielle Bertier; Daniel G MacArthur; Monkol Lek; Esther Lizano; Henk P J Buermans; Ismael Padioleau; Thomas Schwarzmayr; Olof Karlberg; Halit Ongen; Helena Kilpinen; Sergi Beltran; Marta Gut; Katja Kahlem; Vyacheslav Amstislavskiy; Oliver Stegle; Matti Pirinen; Stephen B Montgomery; Peter Donnelly; Mark I McCarthy; Paul Flicek; Tim M Strom; Hans Lehrach; Stefan Schreiber; Ralf Sudbrak; Angel Carracedo; Stylianos E Antonarakis; Robert Häsler; Ann-Christine Syvänen; Gert-Jan van Ommen; Alvis Brazma; Thomas Meitinger; Philip Rosenstiel; Roderic Guigó; Ivo G Gut; Xavier Estivill; Emmanouil T Dermitzakis Journal: Nature Date: 2013-09-15 Impact factor: 49.962
Authors: Jacob D Washburn; Maria Katherine Mejia-Guerra; Guillaume Ramstein; Karl A Kremling; Ravi Valluru; Edward S Buckler; Hai Wang Journal: Proc Natl Acad Sci U S A Date: 2019-03-06 Impact factor: 11.205
Authors: Vahagn C Nikolian; Isabel S Dennahy; Gerald A Higgins; Aaron M Williams; Michael Weykamp; Patrick E Georgoff; Hassan Eidy; Mohamed H Ghandour; Panpan Chang; Hasan B Alam Journal: J Trauma Acute Care Surg Date: 2018-03 Impact factor: 3.313