Predicting drug response of tumors from integrated genomic profiles by deep neural networks

Yu-Chiao Chiu, Hung-I Harry Chen, Tinghe Zhang and 5 more

BMC Medical Genomics · 2019

Abstract

Background. The study of high-throughput genomic profiles from a pharmacogenomics viewpoint has provided unprecedented insights into the oncogenic features modulating drug response. A recent study screened for the response of a thousand human cancer cell lines to a wide collection of anti-cancer drugs and illuminated the link between cellular genotypes and vulnerability. However, due to essential differences between cell lines and tumors, to date the translation into predicting drug response in tumors remains challenging. Recently, advances in deep learning have revolutionized bioinformatics and introduced new techniques to the integration of genomic data. Its application on pharmacogenomics may fill the gap between genomics and drug response and improve the prediction of drug response in tumors.

Results. We proposed a deep learning model to predict drug response (DeepDR) based on mutation and expression profiles of a cancer cell or a tumor. The model contains three deep neural networks (DNNs), i) a mutation encoder pre-trained using a large pan-cancer dataset (The Cancer Genome Atlas; TCGA) to abstract core representations of high-dimension mutation data, ii) a pre-trained expression encoder, and iii) a drug response predictor network integrating the first two subnetworks. Given a pair of mutation and expression profiles, the model predicts IC50 values of 265 drugs. We trained and tested the model on a dataset of 622 cancer cell lines and achieved an overall prediction performance of mean squared error at 1.96 (log-scale IC50 values). The performance was superior in prediction error or stability than two classical methods (linear regression and support vector machine) and four analog DNN models of DeepDR, including DNNs built without TCGA pre-training, partly replaced by principal components, and built on individual types of input data. We then applied the model to predict drug response of 9059 tumors of 33 cancer types. Using per-cancer and pan-cancer settings, the model predicted both known, including EGFR inhibitors in non-small cell lung cancer and tamoxifen in ER+ breast cancer, and novel drug targets, such as vinorelbine for TTN-mutated tumors. The comprehensive analysis further revealed the molecular mechanisms underlying the resistance to a chemotherapeutic drug docetaxel in a pan-cancer setting and the anti-cancer potential of a novel agent, CX-5461, in treating gliomas and hematopoietic malignancies.

Conclusions. Here we present, as far as we know, the first DNN model to translate pharmacogenomics features identified from in vitro drug screening to predict the response of tumors. The results covered both well-studied and novel mechanisms of drug resistance and drug targets. Our model and findings improve the prediction of drug response and the identification of novel therapeutic options.

Full text

Its licence does not allow us to show the full text. Read it at PubMed Central:

References 41

1.
Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646-674.
Not in our library10.1016/j.cell.2011.02.013
2.
Schmitt MW, Loeb LA, Salk JJ. The influence of subclonal resistance mutations on targeted cancer therapy. Nat Rev Clin Oncol. 2016;13(6):335-347.
Not in our library10.1038/nrclinonc.2015.175
3.
Phillips KA, Veenstra DL, Oren E, Lee JK, Sadee W. Potential role of pharmacogenomics in reducing adverse drug reactions: a systematic review. JAMA: the journal of the American Medical Association. 2001;286(18):2270-2279.
Not in our library10.1001/jama.286.18.2270
4.
Hertz DL, Rae J. Pharmacogenetics of cancer drugs. Annu Rev Med. 2015;66:65-81.
Not in our library10.1146/annurev-med-053013-053944
5.
Mina M, Raynaud F, Tavernari D, Battistello E, Sungalee S, Saghafinia S, Laessle T, Sanchez-Vega F, Schultz N, Oricchio E, et al. Conditional selection of genomic alterations dictates Cancer evolution and oncogenic dependencies. Cancer Cell. 2017;32(2):155-168.
Not in our library10.1016/j.ccell.2017.06.010
6.
Park S, Lehner B. Cancer type-dependent genetic interactions between cancer driver alterations indicate plasticity of epistasis across cell types. Mol Syst Biol. 2015;11(7):824.
Not in our library10.15252/msb.20156102
7.
A Landscape of Pharmacogenomic Interactions in Cancer · 2016
No notice
8.
Yang W, Soares J, Greninger P, Edelman EJ, Lightfoot H, Forbes S, Bindal N, Beare D, Smith JA, Thompson IR, et al. Genomics of drug sensitivity in Cancer (GDSC): a resource for therapeutic biomarker discovery in cancer cells. Nucleic Acids Res. 2013;41(Database issue):D955-D961.
Not in our library10.1093/nar/gks1111
9.
Alipanahi B, Delong A, Weirauch MT, Frey BJ. Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning. Nat Biotechnol. 2015;33(8):831-838.
Not in our library10.1038/nbt.3300
10.
Zhou J, Troyanskaya OG. Predicting effects of noncoding variants with deep learning-based sequence model. Nat Methods. 2015;12(10):931-934.
Not in our library10.1038/nmeth.3547
11.
Lin C, Jain S, Kim H, Bar-Joseph Z. Using neural networks for reducing the dimensions of single-cell RNA-Seq data. Nucleic Acids Res. 2017;45(17):e156.
Not in our library10.1093/nar/gkx681
12.
DeepCpG: accurate prediction of single-cell DNA methylation states using deep learning · 2017
Corrected
13.
Chaudhary K, Poirion OB, Lu L, Garmire LX. Deep learning-based multi-omics integration robustly predicts survival in liver Cancer. Clinical cancer research: an official journal of the American Association for Cancer Research. 2018;24(6):1248–59.
Not in our library10.1158/1078-0432.ccr-17-0853
14.
Wen M, Zhang Z, Niu S, Sha H, Yang R, Yun Y, Lu H. Deep-learning-based drug-target interaction prediction. J Proteome Res. 2017;16(4):1401-1409.
Not in our library10.1021/acs.jproteome.6b00618
15.
The cornucopia of meaningful leads: Applying deep adversarial autoencoders for new molecule development in oncology · 2016
No notice
16.
DeepSynergy: predicting anti-cancer drug synergy with Deep Learning · 2017
No notice
17.
Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C. Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods. 2017;14(4):417-419.
Not in our library10.1038/nmeth.4197
18.
Newton Y, Novak AM, Swatloski T, McColl DC, Chopra S, Graim K, Weinstein AS, Baertsch R, Salama SR, Ellrott K, et al. TumorMap: exploring the molecular similarities of Cancer samples in an interactive portal. Cancer Res. 2017;77(21):e111-e114.
Not in our library10.1158/0008-5472.can-17-0580
19.
Barretina J, Caponigro G, Stransky N, Venkatesan K, Margolin AA, Kim S, Wilson CJ, Lehar J, Kryukov GV, Sonkin D, et al. The Cancer cell line encyclopedia enables predictive modelling of anticancer drug sensitivity. Nature. 2012;483(7391):603-607.
Not in our library10.1038/nature11003
20.
Cancer Cell Line Encyclopedia C. Genomics of drug sensitivity in Cancer C: Pharmacogenomic agreement between two cancer cell line data sets. Nature. 2015;528(7580):84-87.
Not in our library10.1038/nature15736
21.
Kowarik A, Templ M. Imputation with the R Package VIM. Journal of Statistical Software. 2016;74(7):1–16.
Not in our library
22.
Alfons A, Templ M. Estimation of Social Exclusion Indicators from Complex Surveys. The R Package laeken 2013. 2013;54(15):25.
Not in our library
23.
He K, Zhang X, Ren S, Sun J. Delving deep into rectifiers: surpassing human-level performance on imagenet classification. Proceedings of the IEEE international conference on computer vision. 2015;2015:1026-1034.
Not in our library
24.
Pumperla M. Keras + Hyperopt: A very simple wrapper for convenient hyperparameter optimization. 2016.
Not in our library
25.
Shetzer Y, Solomon H, Koifman G, Molchadsky A, Horesh S, Rotter V. The paradigm of mutant p53-expressing cancer stem cells and drug resistance. Carcinogenesis. 2014;35(6):1196-1208.
Not in our library10.1093/carcin/bgu073
26.
The role of p53 in cancer drug resistance and targeted chemotherapy · 2016
No notice
27.
Wang JB, Dong DF, Wang MD, Gao K. IDH1 overexpression induced chemotherapy resistance and IDH1 mutation enhanced chemotherapy sensitivity in glioma cells in vitro and in vivo. Asian Pac J Cancer Prev. 2014;15(1):427-432.
Not in our library10.7314/apjcp.2014.15.1.427
28.
Bywater MJ, Poortinga G, Sanij E, Hein N, Peck A, Cullinane C, Wall M, Cluse L, Drygin D, Anderes K, et al. Inhibition of RNA polymerase I as a therapeutic strategy to promote cancer-specific activation of p53. Cancer Cell. 2012;22(1):51-65.
Not in our library10.1016/j.ccr.2012.05.019
29.
Hein N, Cameron DP, Hannan KM, Nguyen NN, Fong CY, Sornkom J, Wall M, Pavy M, Cullinane C, Diesch J, et al. Inhibition of pol I transcription treats murine and human AML by targeting the leukemia-initiating cell population. Blood. 2017;129(21):2882-2895.
Not in our library10.1182/blood-2016-05-718171
30.
Wei JC, Meng FD, Qu K, Wang ZX, Wu QF, Zhang LQ, Pang Q, Liu C. Sorafenib inhibits proliferation and invasion of human hepatocellular carcinoma cells via up-regulation of p53 and suppressing FoxM1. Acta Pharmacol Sin. 2015;36(2):241-251.
Not in our library10.1038/aps.2014.122
31.
Ling X, Calinski D, Chanan-Khan AA, Zhou M, Li F. Cancer cell sensitivity to bortezomib is associated with survivin expression and p53 status but not cancer cell types. J Exp Clin Cancer Res. 2010;29:8.
Not in our library10.1186/1756-9966-29-8
32.
Li P, Zhao M, Parris AB, Feng X, Yang X. p53 is required for metformin-induced growth inhibition, senescence and apoptosis in breast cancer cells. Biochem Biophys Res Commun. 2015;464(4):1267-1274.
Not in our library10.1016/j.bbrc.2015.07.117
33.
Huang d W, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc. 2009;4(1):44-57.
Not in our library10.1038/nprot.2008.211
34.
Huang d W, Sherman BT, Lempicki RA. Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res. 2009;37(1):1-13.
Not in our library10.1093/nar/gkn923
35.
Fulton B, Spencer CM. Docetaxel. A review of its pharmacodynamic and pharmacokinetic properties and therapeutic efficacy in the management of metastatic breast cancer. Drugs. 1996;51(6):1075-1092.
Not in our library10.2165/00003495-199651060-00011
36.
Drygin D, Lin A, Bliesath J, Ho CB, O'Brien SE, Proffitt C, Omori M, Haddach M, Schwaebe MK, Siddiqui-Jain A, et al. Targeting RNA polymerase I with an oral small molecule CX-5461 inhibits ribosomal RNA synthesis and solid tumor growth. Cancer Res. 2011;71(4):1418-1430.
Not in our library10.1158/0008-5472.can-10-1728
37.
CX-5461 is a DNA G-quadruplex stabilizer with selective lethality in BRCA1/2 deficient tumours · 2017
No notice
38.
Deep learning for computational biology · 2016
No notice
39.
Yosinski J, Clune J, Nguyen A, Fuchs T, Lipson H. Understanding neural networks through deep visualization. arXiv preprint arXiv:150606579. 2015.
Not in our library
40.
Shrikumar A, Greenside P, Kundaje A. Learning important features through propagating activation differences. arXiv preprint arXiv:170402685. 2017.
Not in our library

All 41 references

Cited by papers we hold 1

Drug ranking using machine learning systematically predicts the efficacy of anti-cancer drugs · 2021
Not checked

Reads

Read by people on this page

None yet.

Checked or read by AI through naulon

None yet.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.

Are you an author?

Is this your paper?

Claim it to add your note and be paid when agents cite it. Free.