Discovering Synergistic Drug Combinations for COVID with Biological Bottleneck Models

Drug combinations play an important role in therapeutics due to its better efficacy and reduced toxicity. Recent approaches have applied machine learning to identify synergistic combinations for cancer, but they are not applicable to new diseases with limited combination data. Given that drug synergy is closely tied to biological targets, we propose a \emph{biological bottleneck} model that jointly learns drug-target interaction and synergy. The model consists of two parts: a drug-target interaction and target-disease association module. This design enables the model to \emph{explain} how a biological target affects drug synergy. By utilizing additional biological information, our model achieves 0.78 test AUC in drug synergy prediction using only 90 COVID drug combinations for training. We experimentally tested the model predictions in the U.S. National Center for Advancing Translational Sciences (NCATS) facilities and discovered two novel drug combinations (Remdesivir + Reserpine and Remdesivir + IQ-1S) with strong synergy in vitro.

[1]  Regina Barzilay,et al.  Adaptive Invariance for Molecule Property Prediction , 2020, ArXiv.

[2]  Yair Benita,et al.  An Unbiased Oncology Compound Screen to Identify Novel Combination Strategies , 2016, Molecular Cancer Therapeutics.

[3]  Fangfang Xia,et al.  Predicting tumor cell line response to drug pairs with deep learning , 2018, BMC Bioinformatics.

[4]  G. Herrler,et al.  SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor , 2020, Cell.

[5]  Murat Cokol,et al.  Design of high-order antibiotic combinations against M. tuberculosis by ranking and exclusion , 2019, Scientific Reports.

[6]  Albert-László Barabási,et al.  Network-based prediction of drug combinations , 2019, Nature Communications.

[7]  Samuel S. Schoenholz,et al.  Neural Message Passing for Quantum Chemistry , 2017, ICML.

[8]  C. I. Bliss.  THE TOXICITY OF POISONS APPLIED JOINTLY1 , 1939 .

[9]  Marinka Zitnik,et al.  Network Medicine Framework for Identifying Drug Repurposing Opportunities for COVID-19 , 2020, ArXiv.

[10]  F. Cheng,et al.  Network-based drug repurposing for novel coronavirus 2019-nCoV/SARS-CoV-2 , 2020, Cell Discovery.

[11]  Peter J. Park,et al.  Systematic Identification of Synergistic Drug Pairs Targeting HIV , 2012, Nature Biotechnology.

[12]  Regina Barzilay,et al.  Analyzing Learned Molecular Representations for Property Prediction , 2019, J. Chem. Inf. Model..

[13]  Eytan Ruppin,et al.  Discovery of SARS-CoV-2 Antivirals through Large-scale Drug Repositioning , 2020, Nature.

[14]  George Papadatos,et al.  The ChEMBL database in 2017 , 2016, Nucleic Acids Res..

[15]  Andreas Bender,et al.  DeepSynergy: predicting anti-cancer drug synergy with Deep Learning , 2017, Bioinform..

[16]  Benjamin J. Polacco,et al.  A SARS-CoV-2 Protein Interaction Map Reveals Targets for Drug-Repurposing , 2020, Nature.

[17]  Alán Aspuru-Guzik,et al.  Convolutional Networks on Graphs for Learning Molecular Fingerprints , 2015, NIPS.

[18]  S. Loewe,et al.  Die quantitativen Probleme der Pharmakologie , 1928 .

[19]  Sam Michael,et al.  Discovery of Synergistic and Antagonistic Drug Combinations against SARS-CoV-2 In Vitro , 2020, bioRxiv.

[20]  Larry Rubinstein,et al.  The National Cancer Institute ALMANAC: A Comprehensive Screening Resource for the Detection of Anticancer Drug Pairs with Enhanced Therapeutic Activity. , 2017, Cancer research.

[21]  Pedro J. Ballester,et al.  Predicting Synergism of Cancer Drug Combinations Using NCI-ALMANAC Data , 2018, bioRxiv.