The Promise of AI for DILI Prediction

Drug-induced liver injury (DILI) is a common reason for the withdrawal of a drug from the market. Early assessment of DILI risk is an essential part of drug development, but it is rendered challenging prior to clinical trials by the complex factors that give rise to liver damage. Artificial intelligence (AI) approaches, particularly those building on machine learning, range from random forests to more recent techniques such as deep learning, and provide tools that can analyze chemical compounds and accurately predict some of their properties based purely on their structure. This article reviews existing AI approaches to predicting DILI and elaborates on the challenges that arise from the as yet limited availability of data. Future directions are discussed focusing on rich data modalities, such as 3D spheroids, and the slow but steady increase in drugs annotated with DILI risk labels.

[1]  H. L. Morgan.  The Generation of a Unique Machine Description for Chemical Structures-A Technique Developed at Chemical Abstracts Service. , 1965 .

[2]  David Weininger,et al.  SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules , 1988, J. Chem. Inf. Comput. Sci..

[3]  D A Kessler,et al.  Introducing MEDWatch. A new approach to reporting medication and device adverse effects and product problems. , 1993, General hospital psychiatry.

[4]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[5]  Albert P. Li,et al.  A review of the common properties of drugs with idiosyncratic hepatotoxicity and the "multiple determinant hypothesis" for the manifestation of idiosyncratic drug toxicity. , 2002, Chemico-biological interactions.

[6]  Y. Martin,et al.  Do structurally similar molecules have similar biological activity? , 2002, Journal of medicinal chemistry.

[7]  Weida Tong,et al.  Decision Forest: Combining the Predictions of Multiple Independent Decision Tree Models , 2003, J. Chem. Inf. Comput. Sci..

[8]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[9]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[10]  Vladimir Vapnik,et al.  Support-vector networks , 2004, Machine Learning.

[11]  Robert C. Holte,et al.  Very Simple Classification Rules Perform Well on Most Commonly Used Datasets , 1993, Machine Learning.

[12]  Paul A Clemons,et al.  The Connectivity Map: Using Gene-Expression Signatures to Connect Small Molecules, Genes, and Disease , 2006, Science.

[13]  A. Bender,et al.  Circular fingerprints: flexible molecular descriptors with applications from physical chemistry to ADME. , 2006, IDrugs : the investigational drugs journal.

[14]  R. Shoemaker.  The NCI60 human tumour cell line anticancer drug screen , 2006, Nature Reviews Cancer.

[15]  Manuela Pavan,et al.  DRAGON SOFTWARE: AN EASY APPROACH TO MOLECULAR DESCRIPTOR CALCULATIONS , 2006 .

[16]  P. Bernardi,et al.  High concordance of drug-induced human hepatotoxicity with in vitro cytotoxicity measured in a novel cell-based model using high content screening , 2006, Archives of Toxicology.

[17]  Gerta Rücker,et al.  y-Randomization and Its Variants in QSPR/QSAR , 2007, J. Chem. Inf. Model..

[18]  M. Natália D. S. Cordeiro,et al.  Computational chemistry approach for the early detection of drug‐induced idiosyncratic liver toxicity , 2008, J. Comput. Chem..

[19]  N. Kaplowitz,et al.  Drug-induced liver injury. , 2004, Clinical infectious diseases : an official publication of the Infectious Diseases Society of America.

[20]  J. Senior.  Drug hepatotoxicity from a regulatory perspective. , 2007, Clinics in liver disease.

[21]  Peter V. Henstock,et al.  Cellular imaging predictions of clinical drug-induced liver injury. , 2008, Toxicological sciences : an official journal of the Society of Toxicology.

[22]  Weida Tong,et al.  Mold2, Molecular Descriptors from 2D Structures for Chemoinformatics and Toxicoinformatics , 2008, J. Chem. Inf. Model..

[23]  M. Lindquist.  VigiBase, the WHO Global ICSR Database System: Basic Facts , 2008 .

[24]  L. Hood,et al.  Circulating microRNAs, potential biomarkers for drug-induced liver injury , 2009, Proceedings of the National Academy of Sciences.

[25]  Li Fei-Fei,et al.  ImageNet: A large-scale hierarchical image database , 2009, CVPR.

[26]  R. Fontana,et al.  Drug-Induced Liver Injury Network (DILIN) Prospective Study , 2009, Drug safety.

[27]  H Fang,et al.  Genomic indicators in the blood predict drug-induced liver injury , 2010, The Pharmacogenomics Journal.

[28]  Sean Ekins,et al.  A Predictive Ligand-Based Bayesian Model for Human Drug-Induced Liver Injury , 2010, Drug Metabolism and Disposition.

[29]  David Rogers,et al.  Extended-Connectivity Fingerprints , 2010, J. Chem. Inf. Model..

[30]  H. Yamada,et al.  The Japanese toxicogenomics project: application of toxicogenomics. , 2010, Molecular nutrition & food research.

[31]  William M. Lee,et al.  Drugs Associated with Hepatotoxicity and their Reporting Frequency of Liver Adverse Events in VigiBase™ , 2010, Drug safety.

[32]  Nigel Greene,et al.  Developing structure-activity relationships for the prediction of hepatotoxicity. , 2010, Chemical research in toxicology.

[33]  Tudor I. Oprea,et al.  BDDCS Applied to Over 900 Drugs , 2011, The AAPS Journal.

[34]  Yoshua Bengio,et al.  Shallow vs. Deep Sum-Product Networks , 2011, NIPS.

[35]  Chin Yee Liew,et al.  Mixed learning algorithms and features ensemble in hepatotoxicity prediction , 2011, J. Comput. Aided Mol. Des..

[36]  Weida Tong,et al.  FDA-approved drug labeling for the study of drug-induced liver injury. , 2011, Drug discovery today.

[37]  D. Ingber,et al.  From 3D cell culture to organs-on-chips. , 2011, Trends in cell biology.

[38]  CHUN WEI YAP,et al.  PaDEL‐descriptor: An open source software to calculate molecular descriptors and fingerprints , 2011, J. Comput. Chem..

[39]  Weida Tong,et al.  Translating Clinical Findings into Knowledge in Drug Safety Evaluation - Drug Induced Liver Injury Prediction System (DILIps) , 2011, PLoS Comput. Biol..

[40]  Kevin P. Murphy,et al.  Machine learning - a probabilistic perspective , 2012, Adaptive computation and machine learning series.

[41]  J. Ambroso,et al.  Preclinical strategy to reduce clinical hepatotoxicity using in vitro bioactivation data for >200 compounds. , 2012, Chemical research in toxicology.

[42]  Katarzyna R Przybylak,et al.  In silico models for drug-induced liver injury – current status , 2012, Expert opinion on drug metabolism & toxicology.

[43]  J. Kelm,et al.  Multi-cell type human liver microtissues for hepatotoxicity testing , 2012, Archives of Toxicology.

[44]  Hinrich W. H. Göhlmann,et al.  Exploiting the Japanese Toxicogenomics Project for Predictive Modelling of Drug Toxicity , 2012 .

[45]  Quoc V. Le,et al.  Recurrent Neural Networks for Noise Reduction in Robust ASR , 2012, INTERSPEECH.

[46]  Weida Tong,et al.  Quantitative structure-activity relationship models for predicting drug-induced liver injury based on FDA-approved drug labeling annotation and using a large collection of drugs. , 2013, Toxicological sciences : an official journal of the Society of Toxicology.

[47]  J. Serrano,et al.  LiverTox: A website on drug‐induced liver injury , 2013, Hepatology.

[48]  Pierre Baldi,et al.  Deep Architectures and Deep Learning in Chemoinformatics: The Prediction of Aqueous Solubility for Drug-Like Molecules , 2013, J. Chem. Inf. Model..

[49]  Weida Tong,et al.  High lipophilicity and high daily dose of oral medications are associated with significant risk for drug‐induced liver injury , 2013, Hepatology.

[50]  Elizabeth M. Lancaster,et al.  Acetaminophen hepatotoxicity: an updated review , 2014, Archives of Toxicology.

[51]  Xiangwei Zhu,et al.  Hybrid in silico models for drug‐induced liver injury using chemical descriptors and in vitro cell‐imaging information , 2014, Journal of applied toxicology : JAT.

[52]  Naomi L Kruhlak,et al.  Construction and analysis of a human hepatotoxicity database suitable for QSAR modeling using post-market safety data. , 2014, Toxicology.

[53]  Alison J. Foster,et al.  Evaluation of the use of imaging parameters for the detection of compound-induced hepatotoxicity in 384-well cultures of HepG2 cells and cryopreserved primary human hepatocytes. , 2014, Toxicology in vitro : an international journal published in association with BIBRA.

[54]  Weida Tong,et al.  Toward predictive models for drug-induced liver injury in humans: are we there yet? , 2014, Biomarkers in medicine.

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

[56]  Luhua Lai,et al.  Deep Learning for Drug-Induced Liver Injury , 2015, J. Chem. Inf. Model..

[57]  Bie M. P. Verbist,et al.  Using transcriptomics to guide lead optimization in drug discovery projects: Lessons learned from the QSTAR project. , 2015, Drug discovery today.

[58]  William M. Lee,et al.  Features and Outcomes of 899 Patients With Drug-Induced Liver Injury: The DILIN Prospective Study. , 2015, Gastroenterology.

[59]  Robert Preissner,et al.  WITHDRAWN—a resource for withdrawn and discontinued drugs , 2015, Nucleic Acids Res..

[60]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[61]  E. Björnsson,et al.  Categorization of drugs implicated in causing liver injury: Critical assessment based on published case reports , 2016, Hepatology.

[62]  Feixiong Cheng,et al.  In silico Prediction of Drug Induced Liver Toxicity Using Substructure Pattern Recognition Method , 2016, Molecular informatics.

[63]  Weida Tong,et al.  DILIrank: the largest reference drug list ranked by the risk for developing drug-induced liver injury in humans. , 2016, Drug discovery today.

[64]  Simon Messner,et al.  Utility of spherical human liver microtissues for prediction of clinical drug-induced liver injury , 2017, Archives of Toxicology.

[65]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[66]  Maurice Whelan,et al.  How Adverse Outcome Pathways Can Aid the Development and Use of Computational Prediction Models for Regulatory Toxicology , 2016, Toxicological sciences : an official journal of the Society of Toxicology.

[67]  Gerhard F. Ecker,et al.  Predicting drug-induced liver injury: The importance of data curation , 2017, Toxicology.

[68]  Angela N. Brooks,et al.  A Next Generation Connectivity Map: L1000 Platform and the First 1,000,000 Profiles , 2017, Cell.

[69]  W. Tong,et al.  Development of Decision Forest Models for Prediction of Drug-Induced Liver Injury in Humans Using A Large Set of FDA-approved Drugs , 2017, Scientific Reports.

[70]  Sebastian Thrun,et al.  Dermatologist-level classification of skin cancer with deep neural networks , 2017, Nature.

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

[72]  Krister Wennerberg,et al.  A transcriptomics data-driven gene space accurately predicts liver cytopathology and drug-induced liver injury , 2017, Nature Communications.

[73]  Qi Zhao,et al.  Predicting Drug-Induced Liver Injury Using Ensemble Learning Methods and Molecular Fingerprints , 2018, Toxicological sciences : an official journal of the Society of Toxicology.

[74]  Hugo Ceulemans,et al.  Large-scale comparison of machine learning methods for drug target prediction on ChEMBL† †Electronic supplementary information (ESI) available: Overview, Data Collection and Clustering, Methods, Results, Appendix. See DOI: 10.1039/c8sc00148k , 2018, Chemical science.

[75]  Sabine U. Vorrink,et al.  Prediction of Drug-Induced Hepatotoxicity Using Long-Term Stable Primary Hepatic 3D Spheroid Cultures in Chemically Defined Conditions , 2018, Toxicological sciences : an official journal of the Society of Toxicology.

[76]  Weida Tong,et al.  Drug-induced liver injury severity and toxicity (DILIst): binary classification of 1303 drugs by human hepatotoxicity. , 2019, Drug discovery today.

[77]  Alison J. Foster,et al.  Predicting Drug-Induced Liver Injury with Bayesian Machine Learning. , 2019, Chemical research in toxicology.

[78]  G. Danan,et al.  Roussel Uclaf Causality Assessment Method for Drug-Induced Liver Injury: Present and Future , 2019, Front. Pharmacol..

[79]  Xin Yan,et al.  DeepChemStable: Chemical Stability Prediction with an Attention-Based Graph Convolution Network , 2019, J. Chem. Inf. Model..

[80]  Jordi Vitrià,et al.  Modelling heterogeneous distributions with an Uncountable Mixture of Asymmetric Laplacians , 2019, NeurIPS.

[81]  Chunlai Feng,et al.  Gene Expression Data Based Deep Learning Model for Accurate Prediction of Drug-Induced Liver Injury in Advance , 2019, J. Chem. Inf. Model..

[82]  E. Björnsson,et al.  Drug-Induced Liver Injury - Types and Phenotypes. , 2019, The New England journal of medicine.

[83]  Tianyuan Ye,et al.  An In Silico Model for Predicting Drug-Induced Hepatotoxicity , 2019, International journal of molecular sciences.

[84]  Munish Puri,et al.  Automated Machine Learning Diagnostic Support System as a Computational Biomarker for Detecting Drug-Induced Liver Injury Patterns in Whole Slide Liver Pathology Images. , 2020, Assay and drug development technologies.

[85]  W. Tong,et al.  Integrating adverse outcome pathways (AOPs) and high throughput in vitro assays for better risk evaluations, a study with drug-induced liver injury (DILI). , 2019, ALTEX.

[86]  Yangyang Wang,et al.  In Silico Prediction of Drug-Induced Liver Injury Based on Ensemble Classifier Method , 2019, International journal of molecular sciences.

[87]  A. D. Rodrigues,et al.  MOVING BEYOND BINARY PREDICTIONS OF HUMAN DRUG-INDUCED LIVER INJURY (DILI) TOWARDS CONTRASTING RELATIVE RISK POTENTIAL. , 2020, Chemical research in toxicology.

[88]  Sean Ekins,et al.  Comparing Machine Learning Algorithms for Predicting Drug-Induced Liver Injury (DILI). , 2020, Molecular pharmaceutics.

[89]  C. Furlanello,et al.  Predictability of drug-induced liver injury by machine learning , 2020, Biology Direct.

[90]  Ting Li,et al.  Deep Learning on High-Throughput Transcriptomics to Predict Drug-Induced Liver Injury , 2020, Frontiers in Bioengineering and Biotechnology.

[91]  W. Tong,et al.  DeepDILI: Deep Learning-Powered Drug-Induced Liver Injury Prediction Using Model-Level Representation. , 2020, Chemical research in toxicology.

[92]  C. Stephens,et al.  Genetic Predisposition to Drug-Induced Liver Injury. , 2020, Clinics in liver disease.