Using artificial intelligence methods to speed up drug discovery

ABSTRACT

Introduction: Drug discovery is the process through which potential new compounds are identified by means of biology, chemistry, and pharmacology. Due to the high complexity of genomic data, AI techniques are increasingly needed to help reduce this and aid the adoption of optimal decisions. Phenotypic prediction is of particular use to drug discovery and precision medicine where sets of genes that predict a given phenotype are determined. Phenotypic prediction is an undetermined problem given that the number of monitored genetic probes markedly exceeds the number of collected samples (from patients). This imbalance creates ambiguity in the characterization of the biological pathways that are responsible for disease development.

Areas covered: In this paper, the authors present AI methodologies that perform a robust deep sampling of altered genetic pathways to locate new therapeutic targets, assist in drug repurposing and speed up and optimize the drug selection process.

Expert opinion: AI is a potential solution to a number of drug discovery problems, though one should, bear in mind that the quality of data predicts the overall quality of the prediction, as in any modeling task in data science. The use of transparent methodologies is crucial, particularly in drug repositioning/repurposing in rare diseases.

KEYWORDS:

• Drug design
• drug discovery
• phenotype prediction
• artificial intelligence

Article highlights

• Phenotype prediction problems are complex and highly undetermined due to the lack of a theoretical model. Computational methods should be able to relate the causes (altered pathways) to the effects (disease development). The idea consists of finding the set of genes that cause a given diseases and match them with compounds that mitigate genes expressions.

• Phenotype prediction problems are highly underdetermined, since the number of monitored genetic probes is always much larger than the number of observed samples (disease samples).

• A robust sampling with classifiers is required. The uncertainty space relative to the classifier ${\mathbf{L}}^{\ast }$, ${\mathrm{M}}_{\mathrm{t}\mathrm{o}\mathrm{l}}=\left\{\mathbf{g}:\mathrm{O}\left(\mathbf{g}\right)<{\mathrm{E}}_{tol}\right\}$, is composed by the sets of high predictive networks with similar predictive accuracy; that is, those sets of genes $\mathbf{g}$ that classify the samples with a prediction error $\mathrm{O}\left(\mathbf{g}\right)$ lower than ${\mathrm{E}}_{tol}$.

• The main idea is to use genomic, proteomic and metabolomics information, in addition to state-of-the-art drug design models to optimally delineate the optimum targets and compounds.

• AI drug repositioning can be carried out by identifying disease-altered pathways and querying in CMAP the compounds that balances the over and under-expressed genetic pathways, such as in the IBM research carried out by Fernández–Martínez JL et al.

• AI-assisted drug design must be capable of (1) robustly sampling the altered pathways, (2) optimally selecting the right compound which maximizes positive effects while minimizing deleterious ones, (3) predicting the positive and negative of a compound given a random patient, (4) correctly connecting the pharmacological profile of a given drugs with clinical trials and (5) quantifying the uncertainty behind the prediction of drug effects.

This box summarizes key points contained in the article.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer Disclosures

A referee is an employee of Sanofi. Peer reviewers on this manuscript have no other relevant financial relationships or otherwise to disclose.

Additional information

Funding

This manuscript has not been funded.