Linking drug target and pathway activation for effective therapy using multi-task learning

Despite the abundance of large-scale molecular and drug-response data, the insights gained about the mechanisms underlying treatment efficacy in cancer has been in general limited. Machine learning algorithms applied to those datasets most often are used to provide predictions without interpretation, or reveal single drug-gene association and fail to derive robust insights. We propose to use Macau, a bayesian multitask multi-relational algorithm to generalize from individual drugs and genes and explore the interactions between the drug targets and signaling pathways’ activation. A typical insight would be: “Activation of pathway Y will confer sensitivity to any drug targeting protein X”. We applied our methodology to the Genomics of Drug Sensitivity in Cancer (GDSC) screening, using gene expression of 990 cancer cell lines, activity scores of 11 signaling pathways derived from the tool PROGENy as cell line input and 228 nominal targets for 265 drugs as drug input. These interactions can guide a tissue-specific combination treatment strategy, for example suggesting to modulate a certain pathway to maximize the drug response for a given tissue. We confirmed in literature drug combination strategies derived from our result for brain, skin and stomach tissues. Such an analysis of interactions across tissues might help target discovery, drug repurposing and patient stratification strategies.

[1]  Bob van de Water,et al.  MEK inhibition induces apoptosis in osteosarcoma cells with constitutive ERK1/2 phosphorylation , 2015, Genes & cancer.

[2]  Joseph A. DiDonato,et al.  Immunosuppression by Glucocorticoids: Inhibition of NF-κB Activity Through Induction of IκB Synthesis , 1995, Science.

[3]  David Meyre,et al.  From big data analysis to personalized medicine for all: challenges and opportunities , 2015, BMC Medical Genomics.

[4]  Edwin Wang,et al.  Signaling network assessment of mutations and copy number variations predict breast cancer subtype-specific drug targets. , 2013, Cell reports.

[5]  J. Sáez-Rodríguez,et al.  Perturbation-response genes reveal signaling footprints in cancer gene expression , 2016, Nature Communications.

[6]  Emanuel J. V. Gonçalves,et al.  A Landscape of Pharmacogenomic Interactions in Cancer , 2016, Cell.

[7]  J. Trédaniel,et al.  Neurotensin and its high affinity receptor 1 as a potential pharmacological target in cancer therapy , 2013, Front. Endocrin..

[8]  J. Tait,et al.  Challenges and opportunities. , 1996, Journal of psychiatric and mental health nursing.

[9]  Julio Saez-Rodriguez,et al.  Machine Learning Prediction of Cancer Cell Sensitivity to Drugs Based on Genomic and Chemical Properties , 2012, PloS one.

[10]  Krister Wennerberg,et al.  Integrative and Personalized QSAR Analysis in Cancer by Kernelized Bayesian Matrix Factorization , 2014, J. Chem. Inf. Model..

[11]  Qiang Yang,et al.  A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.

[12]  Hristo S. Paskov,et al.  Multitask learning improves prediction of cancer drug sensitivity , 2016, Scientific Reports.

[13]  Ivan Babic,et al.  Oncogenic EGFR signaling activates an mTORC2-NF-κB pathway that promotes chemotherapy resistance. , 2011, Cancer discovery.

[14]  Yves Moreau,et al.  Macau: Scalable Bayesian factorization with high-dimensional side information using MCMC , 2017, 2017 IEEE 27th International Workshop on Machine Learning for Signal Processing (MLSP).

[15]  George Papadatos,et al.  Beyond the hype: deep neural networks outperform established methods using a ChEMBL bioactivity benchmark set , 2017, bioRxiv.

[16]  G. Robertson,et al.  Targeting the MAPK pathway in melanoma: why some approaches succeed and other fail. , 2010, Biochemical pharmacology.

[17]  Webster K. Cavenee,et al.  Orthogonal targeting of EGFRvIII expressing glioblastomas through simultaneous EGFR and PLK1 inhibition , 2015, Oncotarget.

[18]  S Marsoni,et al.  Dual MET/EGFR therapy leads to complete response and resistance prevention in a MET-amplified gastroesophageal xenopatient cohort , 2017, Oncogene.

[19]  Isidro Cortes-Ciriano,et al.  Current Trends in Drug Sensitivity Prediction. , 2017, Current pharmaceutical design.

[20]  Adam A. Margolin,et al.  The Cancer Cell Line Encyclopedia enables predictive modeling of anticancer drug sensitivity , 2012, Nature.

[21]  J. Ruland,et al.  Aberrant NF-kappaB signaling in lymphoma: mechanisms, consequences, and therapeutic implications. , 2007, Blood.

[22]  Jian Jin,et al.  Topoisomerase 1 inhibition suppresses inflammatory genes and protects from death by inflammation , 2016, Science.

[23]  Joshua A. Bittker,et al.  Harnessing Connectivity in a Large-Scale Small-Molecule Sensitivity Dataset. , 2015, Cancer discovery.

[24]  M. Karin,et al.  Immunosuppression by glucocorticoids: inhibition of NF-kappa B activity through induction of I kappa B synthesis. , 1995, Science.

[25]  Davide Corà,et al.  VEGF blockade enhances the antitumor effect of BRAFV 600E inhibition , 2016, EMBO molecular medicine.

[26]  Nasser Ghadiri,et al.  A review of network‐based approaches to drug repositioning , 2018, Briefings Bioinform..