Novel Development of Predictive Feature Fingerprints to Identify Chemistry-Based Features for Effective Drug Design of SARS-CoV-2 Target Antagonists and Inhibitors Using Machine Learning

A unique approach to bioactivity and chemical data curation coupled with Random forest analyses has led to a series of target-specific and cross-validated Predictive Feature Fingerprints (PFF) that have high predictability across multiple therapeutic targets and disease stages involved in the SARS-CoV-2 induced COVID-19 pandemic, which include plasma kallikrein, HIV protease, NSP5, NSP12, JAK family and AT-1. The approach was highly accurate in determining the matched target for the different compound sets and suggests that the models could be used for virtual screening of target specific compound libraries. The curation-modeling process was successfully applied to a SARS-CoV-2 phenotypic screen and could be used for predictive bioactivity estimation and prioritization for clinical trial selection, virtual screening of drug libraries for repurposing of drug molecules, and analysis and direction of proprietary datasets.

Download Full-text

Faculty Opinions recommendation of Development of a virtual screening method for identification of "frequent hitters" in compound libraries.

Faculty Opinions – Post-Publication Peer Review of the Biomedical Literature ◽

10.3410/f.1004071.46608 ◽

2002 ◽

Author(s):

Dennis Hall

Keyword(s):

Virtual Screening ◽

Screening Method ◽

Compound Libraries

Download Full-text

Lead Molecule Prediction and Characterization for Designing MERS-CoV 3C-like Protease Inhibitors: An In silico Approach

Current Computer - Aided Drug Design ◽

10.2174/1573409914666180629151906 ◽

2018 ◽

Vol 15 (1) ◽

pp. 82-88 ◽

Cited By ~ 1

Author(s):

Md. Mostafijur Rahman ◽

Md. Bayejid Hosen ◽

M. Zakir Hossain Howlader ◽

Yearul Kabir

Keyword(s):

Binding Energy ◽

Middle East ◽

Virtual Screening ◽

In Silico ◽

Screening Method ◽

Middle East Respiratory Syndrome ◽

Cytochrome P450 Enzymes ◽

Drug Molecules ◽

Essential Enzyme ◽

Reduced Toxicity

Background: 3C-like protease also called the main protease is an essential enzyme for the completion of the life cycle of Middle East Respiratory Syndrome Coronavirus. In our study we predicted compounds which are capable of inhibiting 3C-like protease, and thus inhibit the lifecycle of Middle East Respiratory Syndrome Coronavirus using in silico methods. Methods: Lead like compounds and drug molecules which are capable of inhibiting 3C-like protease was identified by structure-based virtual screening and ligand-based virtual screening method. Further, the compounds were validated through absorption, distribution, metabolism and excretion filtering. Results: Based on binding energy, ADME properties, and toxicology analysis, we finally selected 3 compounds from structure-based virtual screening (ZINC ID: 75121653, 41131653, and 67266079) having binding energy -7.12, -7.1 and -7.08 Kcal/mol, respectively and 5 compounds from ligandbased virtual screening (ZINC ID: 05576502, 47654332, 04829153, 86434515 and 25626324) having binding energy -49.8, -54.9, -65.6, -61.1 and -66.7 Kcal/mol respectively. All these compounds have good ADME profile and reduced toxicity. Among eight compounds, one is soluble in water and remaining 7 compounds are highly soluble in water. All compounds have bioavailability 0.55 on the scale of 0 to 1. Among the 5 compounds from structure-based virtual screening, 2 compounds showed leadlikeness. All the compounds showed no inhibition of cytochrome P450 enzymes, no blood-brain barrier permeability and no toxic structure in medicinal chemistry profile. All the compounds are not a substrate of P-glycoprotein. Our predicted compounds may be capable of inhibiting 3C-like protease but need some further validation in wet lab.

Download Full-text

A fast virtual screening filter for cytochrome P450 3A4 inhibition liability of compound libraries

Quantitative Structure-Activity Relationships ◽

10.1002/1521-3838(200208)21:3<249::aid-qsar249>3.0.co;2-s ◽

2002 ◽

Vol 21 (3) ◽

pp. 249-256 ◽

Cited By ~ 27

Author(s):

Jochen Zuegge ◽

Uli Fechner ◽

Olivier Roche ◽

Neil J. Parrott ◽

Ola Engkvist ◽

...

Keyword(s):

Cytochrome P450 ◽

Virtual Screening ◽

Cytochrome P450 3A4 ◽

Compound Libraries

Download Full-text

pH Dependent Differential Binding Behavior of Prtotease Inhibitor Molecular Drugs for SARS-COV-2

10.26434/chemrxiv.12009018 ◽

2020 ◽

Author(s):

Hamdullah Khadim Sheikh ◽

Tanzila Arshad ◽

Zainab Sher Mohammad ◽

Iqra Arshad ◽

Mohtasheemul Hassan

Keyword(s):

Intracellular Ph ◽

Binding Energies ◽

Hiv Protease ◽

Binding Characteristics ◽

Binding Behavior ◽

Drug Molecules ◽

Main Protease ◽

Differential Binding ◽

Potent Drug ◽

Inhibitor Drug

In this research we used the structure of SARS-CoV-2 main protease (Mpro) for docking with Anti-HIV protease inhibitor drug molecules within pH 4-8. By carrying out the variance analysis of binding energies at pH 4-8, it was revealed that the binding energy and mode of interaction of the potential ligands with SARS-CoV-2 Mpro, was dependent on variation of pH. We found out that two of the selected protease inhibitors have differential binding characteristics with changing pH hence their binding energies and mode of interaction depends upon intracellular pH. This differential binding behavior can lead to development of pH selective potent drug molecules for binding with viral protease at lowered intracellular pH of virus infected cell.

Download Full-text

How to Benchmark Methods for Structure-Based Virtual Screening of Large Compound Libraries

Methods in Molecular Biology - Computational Drug Discovery and Design ◽

10.1007/978-1-61779-465-0_13 ◽

2011 ◽

pp. 187-195 ◽

Cited By ~ 5

Author(s):

Andrew J. Christofferson ◽

Niu Huang

Keyword(s):

Virtual Screening ◽

Compound Libraries ◽

Large Compound

Download Full-text

Virtual Screening as a Technique for PPAR Modulator Discovery

PPAR Research ◽

10.1155/2010/861238 ◽

2010 ◽

Vol 2010 ◽

pp. 1-10 ◽

Cited By ~ 18

Author(s):

Stephanie N. Lewis ◽

Josep Bassaganya-Riera ◽

David R. Bevan

Keyword(s):

Virtual Screening ◽

Therapeutic Potential ◽

Inflammatory Diseases ◽

Effective Means ◽

Binding Activity ◽

Peroxisome Proliferator ◽

Peroxisome Proliferator Activated Receptor ◽

Compound Libraries ◽

Computational Screening ◽

Reporter Assays

Virtual screening (VS) is a discovery technique to identify novel compounds with therapeutic and preventive efficacy against disease. Our current focus is on the in silico screening and discovery of novel peroxisome proliferator-activated receptor-gamma (PPARγ) agonists. It is well recognized that PPARγagonists have therapeutic applications as insulin sensitizers in type 2 diabetes or as anti-inflammatories. VS is a cost- and time-effective means for identifying small molecules that have therapeutic potential. Our long-term goal is to devise computational approaches for testing the PPARγ-binding activity of extensive naturally occurring compound libraries prior to testing agonist activity using ligand-binding and reporter assays. This review summarizes the high potential for obtaining further fundamental understanding of PPARγbiology and development of novel therapies for treating chronic inflammatory diseases through evolution and implementation of computational screening processes for immunotherapeutics in conjunction with experimental methods for calibration and validation of results.

Download Full-text

The impact of compound library size on the performance of scoring functions for structure-based virtual screening

Briefings in Bioinformatics ◽

10.1093/bib/bbaa095 ◽

2020 ◽

Author(s):

Louison Fresnais ◽

Pedro J Ballester

Keyword(s):

Virtual Screening ◽

Fold Increase ◽

Scoring Functions ◽

Compound Library ◽

Library Size ◽

Compound Libraries ◽

Docking Tool ◽

The Difference ◽

High Level ◽

The Impact

Abstract Larger training datasets have been shown to improve the accuracy of machine learning (ML)-based scoring functions (SFs) for structure-based virtual screening (SBVS). In addition, massive test sets for SBVS, known as ultra-large compound libraries, have been demonstrated to enable the fast discovery of selective drug leads with low-nanomolar potency. This proof-of-concept was carried out on two targets using a single docking tool along with its SF. It is thus unclear whether this high level of performance would generalise to other targets, docking tools and SFs. We found that screening a larger compound library results in more potent actives being identified in all six additional targets using a different docking tool along with its classical SF. Furthermore, we established that a way to improve the potency of the retrieved molecules further is to rank them with more accurate ML-based SFs (we found this to be true in four of the six targets; the difference was not significant in the remaining two targets). A 3-fold increase in average hit rate across targets was also achieved by the ML-based SFs. Lastly, we observed that classical and ML-based SFs often find different actives, which supports using both types of SFs on those targets.

Download Full-text