{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,16]],"date-time":"2026-04-16T08:57:11Z","timestamp":1776329831607,"version":"3.50.1"},"reference-count":35,"publisher":"Oxford University Press (OUP)","issue":"12","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2016,6,15]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>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 \u223c1000 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.<\/jats:p>\n                  <jats:p>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.<\/jats:p>\n                  <jats:p>Availability and implementation: D-GEX is available at https:\/\/github.com\/uci-cbcl\/D-GEX.<\/jats:p>\n                  <jats:p>Contact: \u00a0xhx@ics.uci.edu<\/jats:p>\n                  <jats:p>Supplementary information: \u00a0Supplementary data are available at Bioinformatics online.<\/jats:p>","DOI":"10.1093\/bioinformatics\/btw074","type":"journal-article","created":{"date-parts":[[2016,2,14]],"date-time":"2016-02-14T20:09:07Z","timestamp":1455480547000},"page":"1832-1839","source":"Crossref","is-referenced-by-count":369,"title":["Gene expression inference with deep learning"],"prefix":"10.1093","volume":"32","author":[{"given":"Yifei","family":"Chen","sequence":"first","affiliation":[{"name":"1 Department of Computer Science, University of California, Irvine, CA 92697, USA"},{"name":"4 Baidu Research-Big Data Lab, Beijing, 100085, China"}]},{"given":"Yi","family":"Li","sequence":"additional","affiliation":[{"name":"1 Department of Computer Science, University of California, Irvine, CA 92697, USA"}]},{"given":"Rajiv","family":"Narayan","sequence":"additional","affiliation":[{"name":"2 Broad Institute of MIT And Harvard, Cambridge, MA 02142, USA,"}]},{"given":"Aravind","family":"Subramanian","sequence":"additional","affiliation":[{"name":"2 Broad Institute of MIT And Harvard, Cambridge, MA 02142, USA,"}]},{"given":"Xiaohui","family":"Xie","sequence":"additional","affiliation":[{"name":"1 Department of Computer Science, University of California, Irvine, CA 92697, USA"},{"name":"3 Center for Complex Biological Systems, University of California, Irvine, CA 92697, USA"}]}],"member":"286","published-online":{"date-parts":[[2016,2,11]]},"reference":[{"key":"2023020112295701200_btw074-B1","doi-asserted-by":"crossref","first-page":"648","DOI":"10.1126\/science.1262110","article-title":"The genotype-tissue expression (gtex) pilot analysis: multitissue gene regulation in humans","volume":"348","author":"Ardlie","year":"2015","journal-title":"Science"},{"key":"2023020112295701200_btw074-B2","first-page":"2814","author":"Baldi","year":"2013"},{"key":"2023020112295701200_btw074-B3","doi-asserted-by":"crossref","DOI":"10.1038\/ncomms5308","article-title":"Searching for exotic particles in high-energy physics with deep learning","volume":"5","author":"Baldi","year":"2014","journal-title":"Nat. Commun"},{"key":"2023020112295701200_btw074-B4","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/msb4100158","article-title":"How to infer gene networks from expression profiles","volume":"3","author":"Bansal","year":"2007","journal-title":"Mol. Syst. Biol"},{"key":"2023020112295701200_btw074-B5","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1561\/2200000006","article-title":"Learning deep architectures for AI","volume":"2","author":"Bengio","year":"2009","journal-title":"Found. Trends Mach. Learn"},{"key":"2023020112295701200_btw074-B6","doi-asserted-by":"crossref","first-page":"1798","DOI":"10.1109\/TPAMI.2013.50","article-title":"Representation learning: a review and new perspectives","volume":"35","author":"Bengio","year":"2013","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"2023020112295701200_btw074-B7","doi-asserted-by":"crossref","first-page":"509","DOI":"10.1145\/361002.361007","article-title":"Multidimensional binary search trees used for associative searching","volume":"18","author":"Bentley","year":"1975","journal-title":"Commun. ACM"},{"key":"2023020112295701200_btw074-B8","author":"Bergstra","year":"2010"},{"key":"2023020112295701200_btw074-B9","author":"Chen","year":"2014"},{"key":"2023020112295701200_btw074-B10","author":"Ciresan","year":"2012"},{"key":"2023020112295701200_btw074-B11","author":"Coates","year":"2013"},{"key":"2023020112295701200_btw074-B12","doi-asserted-by":"crossref","first-page":"2449","DOI":"10.1093\/bioinformatics\/bts475","article-title":"Deep architectures for protein contact map prediction","volume":"28","author":"Di Lena","year":"2012","journal-title":"Bioinformatics"},{"key":"2023020112295701200_btw074-B13","doi-asserted-by":"crossref","first-page":"207","DOI":"10.1093\/nar\/30.1.207","article-title":"Gene expression omnibus: NCBI gene expression and hybridization array data repository","volume":"30","author":"Edgar","year":"2002","journal-title":"Nucleic Acids Res"},{"key":"2023020112295701200_btw074-B14","author":"Glorot","year":"2010"},{"key":"2023020112295701200_btw074-B15","author":"Goodfellow","year":"2013"},{"key":"2023020112295701200_btw074-B16","doi-asserted-by":"crossref","first-page":"268","DOI":"10.1038\/35066056","article-title":"Computational studies of gene regulatory networks: in numero molecular biology","volume":"2","author":"Hasty","year":"2001","journal-title":"Nat. Rev. Genet"},{"key":"2023020112295701200_btw074-B17","first-page":"926","article-title":"A practical guide to training restricted Boltzmann machines","volume":"9","author":"Hinton","year":"2010","journal-title":"Momentum"},{"key":"2023020112295701200_btw074-B18","doi-asserted-by":"crossref","first-page":"82","DOI":"10.1109\/MSP.2012.2205597","article-title":"Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups","volume":"29","author":"Hinton","year":"2012","journal-title":"IEEE Signal Process. Mag"},{"key":"2023020112295701200_btw074-B19","author":"Hinton","year":"2012"},{"key":"2023020112295701200_btw074-B20","author":"Krizhevsky","year":"2012"},{"key":"2023020112295701200_btw074-B21","doi-asserted-by":"crossref","first-page":"1929","DOI":"10.1126\/science.1132939","article-title":"The connectivity map: using gene-expression signatures to connect small molecules, genes, and disease","volume":"313","author":"Lamb","year":"2006","journal-title":"Science"},{"key":"2023020112295701200_btw074-B22","doi-asserted-by":"crossref","first-page":"506","DOI":"10.1038\/nature12531","article-title":"Transcriptome and genome sequencing uncovers functional variation in humans","volume":"501","author":"Lappalainen","year":"2013","journal-title":"Nature"},{"key":"2023020112295701200_btw074-B23","doi-asserted-by":"crossref","first-page":"i121","DOI":"10.1093\/bioinformatics\/btu277","article-title":"Deep learning of the tissue-regulated splicing code","volume":"30","author":"Leung","year":"2014","journal-title":"Bioinformatics"},{"key":"2023020112295701200_btw074-B24","doi-asserted-by":"crossref","first-page":"580","DOI":"10.1038\/ng.2653","article-title":"The genotype-tissue expression (GTEX) project","volume":"45","author":"Lonsdale","year":"2013","journal-title":"Nat. Genet"},{"key":"2023020112295701200_btw074-B25","doi-asserted-by":"crossref","first-page":"R61","DOI":"10.1186\/gb-2006-7-7-r61","article-title":"A method for high-throughput gene expression signature analysis","volume":"7","author":"Peck","year":"2006","journal-title":"Genome Biol"},{"key":"2023020112295701200_btw074-B26","first-page":"2825","article-title":"Scikit-learn: machine learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res"},{"key":"2023020112295701200_btw074-B27","article-title":"DANN: a deep learning approach for annotating the pathogenicity of genetic variants","author":"Quang","year":"2014","journal-title":"Bioinformatics"},{"key":"2023020112295701200_btw074-B28","first-page":"5","article-title":"Learning representations by back-propagating errors","author":"Rumelhart","year":"1988","journal-title":"Cogn. Model"},{"key":"2023020112295701200_btw074-B29","author":"Socher","year":"2011"},{"key":"2023020112295701200_btw074-B30","first-page":"1929","article-title":"Dropout: a simple way to prevent neural networks from overfitting","volume":"15","author":"Srivastava","year":"2014","journal-title":"J. Mach. Learn. Res"},{"key":"2023020112295701200_btw074-B31","author":"Sutskever","year":"2013"},{"key":"2023020112295701200_btw074-B32","author":"Theil","year":"1958"},{"key":"2023020112295701200_btw074-B33","author":"Vincent","year":"2008"},{"key":"2023020112295701200_btw074-B34","doi-asserted-by":"crossref","first-page":"926","DOI":"10.1038\/nbt.3001","article-title":"The concordance between RNA-Seq and microarray data depends on chemical treatment and transcript abundance","volume":"32","author":"Wang","year":"2014","journal-title":"Nat. Biotechnol"},{"key":"2023020112295701200_btw074-B35","doi-asserted-by":"crossref","first-page":"e82146","DOI":"10.1371\/journal.pone.0082146","article-title":"Low-rank regularization for learning gene expression programs","volume":"8","author":"Ye","year":"2013","journal-title":"PloS One"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/32\/12\/1832\/49021690\/bioinformatics_32_12_1832.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/32\/12\/1832\/49021690\/bioinformatics_32_12_1832.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,2,1]],"date-time":"2023-02-01T17:34:48Z","timestamp":1675272888000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/32\/12\/1832\/1743989"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2016,2,11]]},"references-count":35,"journal-issue":{"issue":"12","published-print":{"date-parts":[[2016,6,15]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btw074","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/034421","asserted-by":"object"}]},"ISSN":["1367-4811","1367-4803"],"issn-type":[{"value":"1367-4811","type":"electronic"},{"value":"1367-4803","type":"print"}],"subject":[],"published-other":{"date-parts":[[2016,6,15]]},"published":{"date-parts":[[2016,2,11]]}}}