Drug discovery research is how new medications are discovered. Historically, drug discovery, design, and development mostly started with identifying active ingredients from traditional medicines or purely by chance. Later, classical pharmacology was used to investigate chemical libraries including small molecules, natural products, or plant extracts, and find those with therapeutic effects. Since human DNA was sequenced, reverse pharmacology has found remedies to existing diseases through modern testing. Disease processes, molecular compound tests, existing treatments with unanticipated effects, Today the steps in drug discovery and development involve screening hits, iterative medicinal chemistry, and optimization of hits to reduce potential drug side effects (increasing affinity and selectivity). Efficacy or potency, metabolic stability (half-life), and oral bioavailability are also improved in these steps of the drug development process.

Drug discovery and development pipelines are long, complex and depend on numerous factors. Machine learning (ML) approaches provide a set of tools that can improve discovery and decision making for well-specified questions with abundant, high-quality data. Opportunities to apply ML occur in all stages of drug discovery. Examples include target validation, identification of prognostic biomarkers and analysis of digital pathology data in clinical trials. Applications have ranged in context and methodology, with some approaches yielding accurate predictions and insights. The challenges of applying ML lie primarily with the lack of interpretability and repeatability of ML-generated results, which may limit their application. In all areas, systematic and comprehensive high-dimensional data still need to be generated. With ongoing efforts to tackle these issues, as well as increasing awareness of the factors needed to validate ML approaches, the application of ML can promote data-driven decision making and has the potential to speed up the process and reduce failure rates in drug discovery and development. A great variety of experimental data, at a chemical, transcriptomic, or genomic-level is available to readily use for drug development. Summarizing the huge amount of biological data at hand into meaningful models, to grasp the full mechanism of diseases, seems harder and harder. However, systems biology and machine learning approaches are continuously enhanced in order to accelerate the path to efficient drug development. We will focus on three significant related and intermingled questions, that can be subject to automation: drug discovery, drug testing, and drug repurposing. Firstly, this review briefly dwells on the current context in drug development. Later, we will review generic machine learning algorithms, and more specifically, we will focus on sequential learning algorithms and recommender systems. These algorithms have also proven themselves useful in other research fields, and are active biomedical fields of research.

In the next section, we give a summary of the fairly new data types that are openly available as of 2019, and that might be useful in regard to drug development challenges, as many datasets can be integrated to training, validation and feature data for the related algorithms. Indeed, what allows ML techniques to really be efficient is publicly available, curated, annotated data. Multiple types of datasets might be relevant with respect to drug development and drug repositioning questions: information about drug candidates, that are, for instance, the chemical structure of the active molecule, disease gene/protein targets, mechanism of action of drugs, but also their documented side effects. One might also be interested in deducing interesting drug candidates by comparing pairs of diseases, of drugs, of protein targets, and applying the principle of “guilt-by-association

Regards,

Elsa

Submit manuscript at www.walshmedicalmedia.com/submissions/advances-pharmacoepidemiology-drug-safety.html or send as an e-mail attachment to the Editorial Office at submissions@walshmedicalmedia.com