Big Data and Artificial Intelligence in Cancer Drug Discovery: Promise, Challenges, and Emerging Opportunities

Abstract

Simple Summary Only 4.1% of potential cancer therapeutics reach the clinic despite taking roughly fourteen years to develop at a cost of more than a billion USD. Large datasets and artificial intelligence (AI) are promising new tools to improve the odds. Cancer drug discovery produces enormous amounts of data related to cancer targets, candidate drugs, and patients. Separate datasets often use different terms for the same concept, so combining them proves harder than collecting them. Big data is described by seven “V” terms used as descriptors, constraints, and outcomes, and we review how researchers apply AI across cancer drug discovery. Here we add another, Vernacular, a property describing how well separate datasets share the principles, terms, data formats, and exchange methods they need to be analyzed together. We close this review with a list of resources to aggregate data and apply AI to cancer drug discovery.