Machine learning approaches to drug response prediction: challenges and recent progress

Abstract

Cancer is a leading cause of death worldwide. Identifying the best treatment using computational models to personalize drug response prediction holds great promise to improve patient’s chances of successful recovery. Unfortunately, the computational task of predicting drug response is very challenging, partially due to the limitations of the available data and partially due to algorithmic shortcomings. The recent advances in deep learning may open a new chapter in the search for computational drug response prediction models and ultimately result in more accurate tools for therapy response. This review provides an overview of the computational challenges and advances in drug response prediction, and focuses on comparing the machine learning techniques to be of utmost practical use for clinicians and machine learning non-experts. The incorporation of new data modalities such as single-cell profiling, along with techniques that rapidly find effective drug combinations will likely be instrumental in improving cancer care.

Transcriptional Regulation of Cancer Immune Checkpoints: Emerging Strategies for Immunotherapy

“…However, mechanisms can still be tested using gene manipulation techniques such as CRISPR, non-coding RNAs, and/or gene overexpression to prove the concepts, encouraging researchers and pharmaceutical companies to design therapeutics for these candidate targets. Rapid growing in the field of computational drug discovery, artificial intelligence, and big pharmacogenomic/proteomic databases and tools [107][108][109][110][111][112][113][114][115] to predict molecular targets, mechanisms of action, drug responses and adverse effects will eventually benefit the pipeline of the master regulator-targeted immunotherapeutic strategies, even though rounds of extensive benchmarking and testing in vitro and in vivo are needed before full potentials of in silico approaches can be unleashed in clinical settings. Implementation of the user-friendly, web-based programs brings a huge opportunity to scientists and clinicians knowledgeable in biology and disease-specific contexts but have less-to-no coding skills.…”

DrugOrchestra: Jointly predicting drug response, targets, and side effects via deep multi-task learning

“…The improvement is most prominent when using drug targets to help the prediction of drug response. Drug target is known to be one of the most important features in a variety of drug response prediction approaches [42][43][44] , partly due to its central role in understanding drug mechanisms. Interestingly, drug response is much less helpful for drug target prediction.…”

A Scalable, Open-Source Implementation of a Large-Scale Mechanistic Model for Single Cell Proliferation and Death Signaling

“…Advances in statistical and machine learning approaches enable (mostly) data-driven exploration and hypothesis generation from big datasets (5)(6)(7)(8). Trained on features of the input dataset(s), such models can be used for, as just a few examples, to predict drug responses (9)(10)(11) or decide tumor type/stage (12)(13)(14)(15). Although transformative, such machine learning and statistical models have shortcomings.…”