Artificial neural networks had their first heyday in molecular informatics and drug discovery approximately two decades ago. Currently, we are witnessing renewed interest in adapting advanced neural network architectures for pharmaceutical research by borrowing from the field of “deep learning”. Compared with some of the other life sciences, their application in drug discovery is still limited. Here, we provide an overview of this emerging field of molecular informatics, present the basic concepts of prominent deep learning methods and offer motivation to explore these techniques for their usefulness in computer-assisted drug discovery and design. We specifically emphasize deep neural networks, restricted Boltzmann machine networks and convolutional networks.

Deep Learning in Drug Discovery.

Virtual screening (VS) is a powerful technique for identifying hit molecules as starting points for medicinal chemistry. The number of methods and softwares which use the ligand and target-based VS approaches is increasing at a rapid pace. What, however, are the real advantages and disadvantages of the VS technology and how applicable is it to drug discovery projects? This review provides a comprehensive appraisal of several VS approaches currently available. In the first part of this work, an overview of the recent progress and advances in both ligand-based VS (LBVS) and structure-based VS (SBVS) strategies highlighting current problems and limitations will be provided. Special emphasis will be given to in silico chemogenomics approaches which utilize annotated ligand-target as well as protein-ligand interaction databases and which could predict or reveal promiscuous binding and polypharmacology, the knowledge of which would help medicinal chemists to design more potent clinical candidates with fewer side effects. In the second part, recent case studies (all published in the last two years) will be discussed where the VS technology has been applied successfully. A critical analysis of these case studies provides a good platform in order to estimate the applicability of various VS strategies in the new lead identification and optimization.

Virtual screening strategies in drug discovery: a critical review.

Artificial intelligence (AI) tools are increasingly being applied in drug discovery. While some protagonists point to vast opportunities potentially offered by such tools, others remain sceptical, waiting for a clear impact to be shown in drug discovery projects. The reality is probably somewhere in-between these extremes, yet it is clear that AI is providing new challenges not only for the scientists involved but also for the biopharma industry and its established processes for discovering and developing new medicines. This article presents the views of a diverse group of international experts on the 'grand challenges' in small-molecule drug discovery with AI and the approaches to address them.

/pdf/rethinking-drug-design-in-the-artificial-intelligence-era-23sseci6dz.pdf

Rethinking drug design in the artificial intelligence era

Structure-based drug design is becoming an essential tool for faster and more cost-efficient lead discovery relative to the traditional method. Genomic, proteomic, and structural studies have provided hundreds of new targets and opportunities for future drug discovery. This situation poses a major problem: the necessity to handle the “big data” generated by combinatorial chemistry. Artificial intelligence (AI) and deep learning play a pivotal role in the analysis and systemization of larger data sets by statistical machine learning methods. Advanced AI-based sophisticated machine learning tools have a significant impact on the drug discovery process including medicinal chemistry. In this review, we focus on the currently available methods and algorithms for structure-based drug design including virtual screening and de novo drug design, with a special emphasis on AI- and deep-learning-based methods used for drug discovery.

https://www.mdpi.com/1422-0067/20/11/2783/pdf

A Structure-Based Drug Discovery Paradigm.

In recent decades, in silico absorption, distribution, metabolism, excretion (ADME), and toxicity (T) modelling as a tool for rational drug design has received considerable attention from pharmaceutical scientists, and various ADME/T-related prediction models have been reported. The high-throughput and low-cost nature of these models permits a more streamlined drug development process in which the identification of hits or their structural optimization can be guided based on a parallel investigation of bioavailability and safety, along with activity. However, the effectiveness of these tools is highly dependent on their capacity to cope with needs at different stages, e.g. their use in candidate selection has been limited due to their lack of the required predictability. For some events or endpoints involving more complex mechanisms, the current in silico approaches still need further improvement. In this review, we will briefly introduce the development of in silico models for some physicochemical parameters, ADME properties and toxicity evaluation, with an emphasis on the modelling approaches thereof, their application in drug discovery, and the potential merits or deficiencies of these models. Finally, the outlook for future ADME/T modelling based on big data analysis and systems sciences will be discussed.

/pdf/in-silico-adme-t-modelling-for-rational-drug-design-3wgzkqxrle.pdf

In silico ADME/T modelling for rational drug design

https://www.elsevier.com/__data/assets/pdf_file/0008/97064/Data_driven_medicinal_chemistry.pdf

Data-driven medicinal chemistry in the era of big data

A pharmacophore describes the arrangement of molecular features a ligand must contain to efficaciously bind a receptor. Pharmacophore models are developed to improve molecular understanding of ligand–protein interactions, and can be used as a tool to identify novel compounds that fulfil the pharmacophore requirements and have a high probability of being biologically active. Protein structure-based pharmacophores (SBPs) derive these molecular features by conversion of protein properties to reciprocal ligand space. Unlike ligand-based pharmacophore models, which require templates of ligands in their bioactive conformation, SBPs do not depend on ligand information. The current review describes the different steps in the construction of SBPs: (i) protein structure preparation, (ii) binding site detection, (iii) pharmacophore feature definition, and (iv) pharmacophore feature selection. We show that the SBP modeling workflow poses different challenges than ligand-based pharmacophore modeling, including the definition of protein pharmacophore features essential for ligand binding. A comprehensive overview of different SBP modeling and screening methods and applications is provided to illustrate that SBPs can be efficiently used for virtual screening, ligand binding mode prediction, and binding site similarity detection. Our review demonstrates that SBPs are valuable tools for hit and lead optimization, compound library design and target hopping, especially in cases where ligand information is scarce.

de J Jacob Vlieg

Papers

Data-driven medicinal chemistry in the era of big data

From the protein's perspective: the benefits and challenges of protein structure-based pharmacophore modeling