Drug target inference by mining transcriptional data using a novel graph convolutional network framework

Abstract

A fundamental challenge that arises in biomedicine is the need to characterize compounds in a relevant cellular context in order to reveal potential on-target or offtarget effects. Recently, the fast accumulation of gene transcriptional profiling data provides us an unprecedented opportunity to explore the protein targets of chemical compounds from the perspective of cell transcriptomics and RNA biology. Here, we propose a novel Siamese spectral-based graph convolutional network (SSGCN) model for inferring the protein targets of chemical compounds from gene transcriptional profiles. Although the gene signature of a compound perturbation only provides indirect clues of the interacting targets, and the biological networks under different experiment conditions further complicate the situation, the SSGCN model was successfully trained to learn from known compound-target pairs by uncovering the hidden correlations between compound perturbation profiles and gene knockdown profiles. On a benchmark set and a large time-split validation dataset, the model achieved higher target inference accuracy as compared to previous methods such as Connectivity Map. Further experimental validations of prediction results highlight the practical usefulness of SSGCN in either inferring the interacting targets of compound, or reversely, in finding novel inhibitors of a given target of interest.

Keywords

drug target inference / transcriptomics / deep learning / experimental verification

References

[1] Abbas AK, Trotta E, Simeonov DR, Marson A, Bluestone JA (2018) Revisiting IL-2: biology and therapeutic prospects. Sci Immunol 3: eaat1482.

[2] André F, Ciruelos E, Rubovszky G, Campone M, Loibl S, Rugo HS, Iwata H, Conte P, Mayer IA, Kaufman B (2019) Alpelisib for PIK3CA-mutated, hormone receptor-positive advanced breast cancer. New Engl J Med. 380: 1929- 1940

[3] Anighoro A, Bajorath J, Rastelli G (2014) Polypharmacology:challenges and opportunities in drug discovery. J Med Chem 57: 7874- 7887

[4] Arshad U, Pertinez H, Box H, Tatham L, Rajoli RKR, Curley P, Neary M, Sharp J, Liptrott NJ, Valentijn A et al (2020) Prioritization of anti-SARS-Cov-2 drug repurposing opportunities based on plasma and target site concentrations derived from their established human pharmacokinetics. Clin Pharmacol Ther. https://doi.org/10.1002/cpt.1909

[5] Ashburn TT, Thor KB (2004) Drug repositioning: Identifying and developing new uses for existing drugs. Nat Rev Drug Discov 3: 673- 683

[6] Bajorath J (2014) Evolution of the activity cliff concept for structureactivity relationship analysis and drug discovery. Future Med Chem 6: 1545- 1549

[7] Behm VY, Blumberg J, Bush C, Grover R, Minich D, Newton R, Perlmutter D, Reed D, Sinatra S, Stroka M (2020) Personalized nutrition & the COVID-19 Era. https://theana.org/COVID-19

[8] Bernardo D, Thompson MJ, Gardner TS, Chobot SE, Eastwood EL, Wojtovich AP, Elliott SJ, Schaus SE, Collins JJ (2005a) Chemogenomic profiling on a genomewide scale using reverseengineered gene networks. Nat Biotechnol 23: 377- 383

[9] Braaten D, Luban J (2001) Cyclophilin A regulates HIV-1 infectivity, as demonstrated by gene targeting in human T cells. EMBO J 20: 1300- 1309

[10] Bruna J (2014) Spectral networks and deep locally connected networks on graphs. https://arxiv.org/abs/1312.6203.

[11] Carozza JA, Böhnert V, Nguyen KC, Skariah G, Shaw KE, Brown JA, Rafat M, von Eyben R, Graves EE, Glenn JS et al (2020) Extracellular cGAMP is a cancer-cell-produced immunotransmitter involved in radiation-induced anticancer immunity. Nat Cancer 1: 184- 196

[12] Cavagna L, Seminari E, Zanframundo G, Gregorini M, Di Matteo A, Rampino T, Montecucco C, Pelenghi S, Cattadori B, Pattonieri EF et al (2020) Calcineurin inhibitor-based immunosuppression and COVID-19: results from a multidisciplinary cohort of patients in Northern Italy. Microorganisms 8: 977

[13] Cereto-Massagué A, Ojeda MJ, Valls C, Mulero M, Pujadas G, Garcia-Vallve S (2015) Tools for in silico target fishing. Methods 71: 98- 103

[14] Chua HN, Roth FP (2011) Discovering the targets of drugs via computational systems biology. J Biol Chem 286: 23653- 23658

[15] Cimmperman P, Baranauskiene L, Jachimoviciūte S, Jachno J, Torresan J, Michailoviene V, Matuliene J, Sereikaite J, Bumelis V, Matulis D (2008) A quantitative model of thermal stabilization and destabilization of proteins by ligands. Biophys J 95: 3222- 3231

[16] Corrales L, Glickman LH, McWhirter SM, Kanne DB, Sivick KE, Katibah GE, Woo SR, Lemmens E, Banda T, Leong JJ et al (2015) Direct activation of STING in the tumor microenvironment leads to potent and systemic tumor regression and immunity. Cell Rep 11: 1018- 1030

[17] Cosgrove EJ, Zhou Y, Gardner TS, Kolaczyk ED (2008) Predicting gene targets of perturbations via network-based filtering of mRNA expression compendia. Bioinformatics 24: 2482- 2490

[18] Dawar FU, Xiong Y, Khattak MNK, Li J, Lin L, Mei J (2017) Potential role of cyclophilin A in regulating cytokine secretion. J Leukoc Biol 102: 989- 992

[19] Enache OM, Lahr DL, Natoli TE, Litichevskiy L, Wadden D, Flynn C, Gould J, Asiedu JK, Narayan R, Subramanian A (2019) The GCTx format and cmap Py, R, M, J packages: resources for optimized storage and integrated traversal of annotated dense matrices. Bioinformatics 35: 1427- 1429

[20] Equils O, Shapiro A, Madak Z, Liu C, Lu D (2004) Human immunodeficiency virus type 1 protease inhibitors block toll-like receptor 2 (TLR2)- and TLR4-Induced NF-kappaB activation. Antimicrob Agents Chemother 48: 3905- 3911

[21] Fedorov O, Marsden B, Pogacic V, Rellos P, Müller S, Bullock AN, Schwaller J, Sundström M, Knapp S (2007) A systematic interaction map of validated kinase inhibitors with Ser/Thr kinases. Proc Natl Acad Sci USA 104: 20523- 20528

[22] Filzen TM, Kutchukian PS, Hermes JD, Li J, Tudor M (2017) Representing high throughput expression profiles via perturbation barcodes reveals compound targets. PLoS Comp Bio 13: e1005335.

[23] Fish PV, Filippakopoulos P, Bish G, Brennan PE, Bunnage ME, Cook AS, Federov O, Gerstenberger BS, Jones H, Knapp S (2012) Identification of a chemical probe for bromo and extra C-terminal bromodomain inhibition through optimization of a fragment-derived hit. J Med Chem 55: 9831- 9837

[24] Gallatin WM, Dietsch GN, Odingo J, Florio V (2019) Ectonucleotide pyrophosphatase-phosphodiesterase (ENPP) inhibitors and uses thereof. (Mavupharma, Inc., USA)

[25] Gardner TS, Di Bernardo D, Lorenz D, Collins JJ (2003) Inferring genetic networks and identifying compound mode of action via expression profiling. Science 301: 102- 105

[26] Geppert H, Vogt M, Bajorath J (2010) Current trends in ligand-based virtual screening: molecular representations, data mining methods, new application areas, and performance evaluation. J Chem Inf Model 50: 205- 216

[27] Hamilton WL, Ying R, Leskovec J (2017) Representation learning on graphs: methods and applications. https://arxiv.org/abs/1709.05584

[28] Hirakawa M, Matos TR, Liu H, Koreth J, Kim HT, Paul NE, Murase K, Whangbo J, Alho AC, Nikiforow S, et al (2016) Low-dose IL-2 selectively activates subsets of CD4+ Tregs and NK cells. JCI Insight 1: e89278.

[29] Huang C, Wang Y, Li X, Ren L, Zhao J, Hu Y, Zhang L, Fan G, Xu J, Gu X et al (2020) Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet 395: 497- 506

[30] Ianevski A, Yao R, Fenstad MH, Biza S, Zusinaite E, Reisberg T, Lysvand H, Løseth K, Landsem VM, Malmring JF et al (2020) Potential antiviral options against SARS-CoV-2 infection. Viruses 12: 642

[31] Iorio F, Bosotti R, Scacheri E, Belcastro V, Mithbaokar P, Ferriero R, Murino L, Tagliaferri R, Brunetti-Pierri N, Isacchi A (2010) Discovery of drug mode of action and drug repositioning from transcriptional responses. Proc Natl Acad Sci USA 107: 14621- 14626

[32] Jackson AL, Bartz SR, Schelter J, Kobayashi SV, Burchard J, Mao M, Li B, Cavet G, Linsley PS (2003) Expression profiling reveals off-target gene regulation by RNAi. Nat Biotechnol 21: 635- 637

[33] Kabir A, Honda RP, Kamatari YO, Endo S, Fukuoka M, Kuwata K (2016) Effects of ligand binding on the stability of aldo-keto reductases: implications for stabilizer or destabilizer chaperones. Protein Sci 25: 2132- 2141

[34] Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. arXiv:1412.6980

[35] Koleti A, Terryn R, Stathias V, Chung C, Cooper DJ, Turner JP, Vidović D, Forlin M, Kelley TT, D’Urso A (2018) Data portal for the library of integrated network-based cellular signatures (LINCS) program: integrated access to diverse large-scale cellular perturbation response data. Nucl Acids Res 46: D558- D566

[36] Lamb J, Crawford ED, Peck D, Modell JW, Blat IC, Wrobel MJ, Lerner J, Brunet JP, Subramanian A, Ross KN et al (2006) The connectivity map: using gene-expression signatures to connect small molecules, genes, and disease. Science 313: 1929- 1935

[37] Leek JT, Scharpf RB, Bravo HC, Simcha D, Langmead B, Johnson WE, Geman D, Baggerly K, Irizarry RA (2010) Tackling the widespread and critical impact of batch effects in high-throughput data. Nat Rev Genet 11: 733- 739

[38] Lenselink EB, Ten Dijke N, Bongers B, Papadatos G, Van Vlijmen HWT, Kowalczyk W, Ijzerman AP, Van Westen GJP (2017) Beyond the hype: deep neural networks outperform established methods using a ChEMBL bioactivity benchmark set. J Cheminform 9: 1- 14

[39] Li L, Yin Q, Kuss P, Maliga Z, Millan JL, Wu H, Mitchison TJ (2014) Hydrolysis of 2’3’-cGAMP by ENPP1 and design of nonhydrolyzable analogs. Nat Chem Biol 10: 1043- 1048

[40] Liang X, Young WC, Hung L-H, Raftery AE, Yeung KY (2019) Integration of multiple data sources for gene network inference using genetic perturbation data. J Comput Biol 26: 1113- 1129

[41] Liu TP, Hsieh YY, Chou CJ, Yang PM (2018) Systematic polypharmacology and drug repurposing via an integrated L1000-based connectivity map database mining. R Soc Open Sci 5: 181321

[42] Madhukar NS, Khade PK, Huang L, Gayvert K, Galletti G, Stogniew M, Allen JE, Giannakakou P, Elemento O (2019) A Bayesian machine learning approach for drug target identification using diverse data types. Nat Commun 10: 5221

[43] Musa A, Ghoraie LS, Zhang SD, Glazko G, Yli-Harja O, Dehmer M, Haibe-Kains B, Emmert-Streib F (2018) A review of connectivity map and computational approaches in pharmacogenomics. Brief Bioinform 19: 506- 523

[44] Musarrat F, Chouljenko V, Dahal A, Nabi R, Chouljenko T, Jois SD, Kousoulas KG (2020) The anti-HIV drug nelfinavir mesylate(Viracept) is a potent inhibitor of cell fusion caused by the SARSCoV-2 spike (S) glycoprotein warranting further evaluation as an antiviral against COVID-19 infections. J Med Virol https://doi.org/10.1002/jmv.25985.

[45] Noh H, Gunawan R (2016) Inferring gene targets of drugs and chemical compounds from gene expression profiles. Bioinformatics 32: 2120- 2127

[46] Noh H, Shoemaker JE, Gunawan R (2018) Network perturbation analysis of gene transcriptional profiles reveals protein targets and mechanism of action of drugs and influenza A viral infection. Nucl Acids Res 46: e34.

[47] Novotny-Diermayr V, Sangthongpitag K, Hu CY, Wu X, Sausgruber N, Yeo P, Greicius G, Pettersson S, Liang AL, Loh YK (2010) SB939, a novel potent and orally active histone deacetylase inhibitor with high tumor exposure and efficacy in mouse models of colorectal cancer. Mol Cancer Ther 9: 642- 652

[48] Ohashi H, Watashi K, Saso W, Shionoya K, Iwanami S, Hirokawa T, Shirai T, Kanaya S, Ito Y, Kim KS, et al (2020) Multidrug treatment with nelfinavir and cepharanthine against COVID-19. https://doi.org/10.1101/2020.04.14.039925v1.

[49] Pabon NA, Xia Y, Estabrooks SK, Ye Z, Herbrand AK, Süß E, Biondi RM, Assimon VA, Gestwicki JE, Brodsky JL, et al (2018) Predicting protein targets for drug-like compounds using transcriptomics. PLOS Commun Biol 14: e1006651.

[50] Pabon NA, Zhang Q, Cruz JA, Schipper DL, Camacho CJ, Lee REC (2019) A network-centric approach to drugging TNF-induced NF-κB signaling. Nat Commun 10: 860

[51] Pacold ME, Brimacombe KR, Chan SH, Rohde JM, Lewis CA, Swier LJYM, Possemato R, Chen WW, Sullivan LB, Fiske BP et al (2016) A PHGDH inhibitor reveals coordination of serine synthesis and one-carbon unit fate. Nat Chem Biol 12: 452- 458

[52] Ramanjulu JM, Pesiridis GS, Yang J, Concha N, Singhaus R, Zhang SY, Tran JL, Moore P, Lehmann S, Eberl HC et al (2018) Design of amidobenzimidazole STING receptor agonists with systemic activity. Nature 564: 439- 443

[53] Salviato E, Djordjilović V, Chiogna M, Romualdi C (2019) SourceSet:a graphical model approach to identify primary genes in perturbed biological pathways. PLoS Comp Biol 15: e1007357.

[54] Schenone M, Dančík V, Wagner BK, Clemons PA (2013) Target identification and mechanism of action in chemical biology and drug discovery. Nat Chem Biol 9: 232- 240

[55] Schomburg KT, Bietz S, Briem H, Henzler AM, Urbaczek S, Rarey M (2014) Facing the challenges of structure-based target prediction by inverse virtual screening. J Chem Inf Model 54: 1676- 1686

[56] Sheridan RP (2013) Time-split cross-validation as a method for estimating the goodness of prospective prediction. J Chem Inf Model 53: 783- 790

[57] Sramek M, Neradil J, Veselska R (2017) Much more than you expected: the non-DHFR-mediated effects of methotrexate. Biochim 1861: 499- 503

[58] Subramanian A, Narayan R, Corsello SM, Peck DD, Natoli TE, Lu XD, Gould J, Davis JF, Tubelli AA, Asiedu JK et al (2017) A next generation connectivity map: L1000 platform and the first 1,000,000 profiles. Cell 171: 1437- 1452

[59] Sun B, Shah B, Fiskus W, Qi J, Rajapakshe K, Coarfa C, Li L, Devaraj SGT, Sharma S, Zhang L et al (2015) Synergistic activity of BET protein antagonist-based combinations in mantle cell lymphoma cells sensitive or resistant to ibrutinib. Blood 126: 1565- 1574

[60] Svensson F, Karlén A, Sköld C (2012) Virtual screening data fusion using both structure- and ligand-based methods. J Chem Inf Model 52: 225- 232

[61] Sydow D, Burggraaff L, Szengel A, van Vlijmen HWT, Ijzerman AP, van Westen GJP, Volkamer A (2019) Advances and challenges in computational target prediction. J Chem Inf Model 59: 1728- 1742

[62] Szklarczyk D, Gable AL, Lyon D, Junge A, Wyder S, Huerta-Cepas J, Simonovic M, Doncheva NT, Morris JH, Bork P (2019) STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucl Acids Res 47: D607- D613

[63] Tanaka Y, Sato Y, Sasaki T (2013) Suppression of coronavirus replication by cyclophilin inhibitors. Viruses 5: 1250- 1260

[64] Terrett NK, Bell AS, Brown D, Ellis P (1996) Sildenafil (VIAGRATM), a potent and selective inhibitor of type 5 cGMP phosphodiesterase with utility for the treatment of male erectile dysfunction. Bioorg Med Chem Lett 6: 1819- 1824

[65] Timme N, Han Y, Liu S, Yosief HO, García HD, Bei Y, Klironomos F, MacArthur IC, Szymansky A, von Stebut JJTO (2020) Smallmolecule dual PLK1 and BRD4 inhibitors are active against preclinical models of pediatric solid tumors. Transl Oncol 13: 221- 232

[66] Wallet MA, Reist CM, Williams JC, Appelberg S, Guiulfo GL, Gardner B, Sleasman JW, Goodenow MM (2012) The HIV-1 protease inhibitor nelfinavir activates PP2 and inhibits MAPK signaling in macrophages: a pathway to reduce inflammation. J Leukoc Biol 92: 795- 805

[67] Wang M, Noh H, Mochan E, Shoemaker JE (2020) Network insights into improving drug target inference algorithms. Preprint at. https://doi.org/10.1101/2020.01.17.910885

[68] Woo JH, Shimoni Y, Yang WS, Subramaniam P, Iyer A, Nicoletti P, Martínez MR, López G, Mattioli M, Realubit R (2015) Elucidating compound mechanism of action by network perturbation analysis. Cell 162: 441- 451

[69] Xie L, He S, Song X, Bo X, Zhang Z (2018) Deep learning-based transcriptome data classification for drug-target interaction prediction. BMC Genomics 19: 667

[70] Xu C, Ai DS, Suo SB, Chen XW, Yan YZ, Cao YQ, Sun N, Chen WZ, McDermott J, Zhang SQ et al (2018) Accurate drug repositioning through non-tissue-specific core signatures from cancer transcriptomes. Cell Rep 25: 523- 535

[71] Xu L, Song X, Su L, Zheng Y, Li R, Sun J (2019) New therapeutic strategies based on IL-2 to modulate Treg cells for autoimmune diseases. Int Immunopharmacol 72: 322- 329

[72] Xu Z, Peng C, Shi Y, Zhu Z, Mu K, Wang X, Zhu W (2020a) Nelfinavir was predicted to be a potential inhibitor of 2019-nCov main protease by an integrative approach combining homology modelling, molecular docking and binding free energy calculation. https://doi.org/10.1101/2020.01.27.921627v1

[73] Xu Z, Yao H, Shen J, Wu N, Xu Y, Lu X, Zhu W, Li L-J (2020b) Nelfinavir is active against SARS-CoV-2 in Vero E6 cells. https://chemrxiv.org/articles/Nelfinavir_Is_Active_Against_SARS-CoV-2_in_Vero_E6_Cells/12039888.

[74] Yamamoto N, Matsuyama S, Hoshino T, Yamamoto N (2020) Nelfinavir inhibits replication of severe acute respiratory syndrome coronavirus 2 in vitro. https://doi.org/10.1101/2020.04.06.026476v1.

[75] Zhao L-H, Zhou XE, Yi W, Wu Z, Liu Y, Kang Y, Hou L, de Waal PW, Li S, Jiang Y et al (2015) Destabilization of strigolactone receptor DWARF14 by binding of ligand and E3-ligase signaling effector DWARF3. Cell Res 25: 1219- 1236

RIGHTS & PERMISSIONS

2021 The Author(s)