Machine learning techniques applied to the drug design and discovery of new antivirals: a brief look over the past decade

ABSTRACT

Introduction: Drug design and discovery of new antivirals will always be extremely important in medicinal chemistry, taking into account known and new viral diseases that are yet to come. Although machine learning (ML) have shown to improve predictions on the biological potential of chemicals and accelerate the discovery of drugs over the past decade, new methods and their combinations have improved their performance and established promising perspectives regarding ML in the search for new antivirals.

Areas covered: The authors consider some interesting areas that deal with different ML techniques applied to antivirals. Recent innovative studies on ML and antivirals were selected and analyzed in detail. Also, the authors provide a brief look at the past to the present to detect advances and bottlenecks in the area.

Expert opinion: From classical ML techniques, it was possible to boost the searches for antivirals. However, from the emergence of new algorithms and the improvement in old approaches, promising results will be achieved every day, as we have observed in the case of SARS-CoV-2. Recent experience has shown that it is possible to use ML to discover new antiviral candidates from virtual screening and drug repurposing.

KEYWORDS:

• Machine learning
• antivirals
• drug design
• drug discovery

Article highlights

• Machine learning (ML) techniques constitute a large and important group of algorithms used to detect, analyze or predict patterns in different datasets, in particular chemical and biological data of interest in medicinal chemistry.

• ML techniques are valuable tools for understanding the chemical and biological mechanisms related to viral diseases.

• ML techniques can be applied individually or combined with other computational approaches. Molecular structures, physicochemical properties, and structure-activity relationships (SAR) of datasets containing antivirals can be evaluated using ML algorithms.

• Recent studies have described the use of ML algorithms (e.g. transfer learning, deep learning, and multi-task learning) for different purposes, such as prediction on experimental conditions related to the synthesis of bioactive substances, construction of predictive models, and analysis of molecular representations and descriptors. These and other applications of ML are interesting topics to be considered in the drug design and discovery of new antivirals.

• The combination of different ML techniques can be useful in improving an existing algorithm or even refining an ML approach. Strategies such as ‘design, make, test and analyze’ (DMTA) and ‘computer-aided synthesis planning’ (CASP) may be of use in this scenario towards perfecting prediction models from ML and reducing their potential failures.

• It is noticeable the growing number of studies applying different ML techniques to design and discover new antivirals, as illustrated from the findings from 2011-2020, as well as the good performance of these methods to predict compounds against different virus targets.

This box summarizes key points contained in the article.

Declaration of interest

The authors have no other 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 apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

KM Honorio is supported by the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), the Conselho Nacional de desenvolvimento Científico e Tecnológico (CNPq), the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) and Pro-Reitoria de Pesquisa, Universidade de São Paulo (PRP-USP). Furthermore MSM Serafim, VS dos Santos Junior, JC Gertrudes and VG Maltarollo are also supported by CAPES. JC Gertrudes and VG Maltarollo are also supported by the CNPq.