The application of machine learning techniques to innovative antibacterial discovery and development

Serafim, Mateus Sá Magalhães; Kronenberger, Thales; Oliveira, Patrícia Rufino; Poso, Antti; Honório, Káthia Maria; Mota, Bruno Eduardo Fernandes

doi:10.1080/17460441.2020.1776696

Cited by 37 publications

(22 citation statements)

References 114 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ligands were filtered using ADMET Predictor version 8.0 [ 40 ] to ensure compliance with the Lipinski’s Rule of five (RO5), which is as follows: compounds with a molecular mass of less than 500 Da, have no more than 5 hydrogen bond donors, no more than 10 hydrogen bond acceptors, and the octanol-water partition coefficient log is no greater than 5 [ 41 , 42 ]. The ADMET Risk model helps to identify any potential liabilities that are likely to impede the success of the prospective drug design.…”

Section: Methodsmentioning

confidence: 99%

Computational Study on Potential Novel Anti-Ebola Virus Protein VP35 Natural Compounds

et al. 2021

View full text Add to dashboard Cite

Ebola virus (EBOV) is one of the most lethal pathogens that can infect humans. The Ebola viral protein VP35 (EBOV VP35) inhibits host IFN-α/β production by interfering with host immune responses to viral invasion and is thus considered as a plausible drug target. The aim of this study was to identify potential novel lead compounds against EBOV VP35 using computational techniques in drug discovery. The 3D structure of the EBOV VP35 with PDB ID: 3FKE was used for molecular docking studies. An integrated library of 7675 African natural product was pre-filtered using ADMET risk, with a threshold of 7 and, as a result, 1470 ligands were obtained for the downstream molecular docking using AutoDock Vina, after an energy minimization of the protein via GROMACS. Five known inhibitors, namely, amodiaquine, chloroquine, gossypetin, taxifolin and EGCG were used as standard control compounds for this study. The area under the curve (AUC) value, evaluating the docking protocol obtained from the receiver operating characteristic (ROC) curve, generated was 0.72, which was considered to be acceptable. The four identified potential lead compounds of NANPDB4048, NANPDB2412, ZINC000095486250 and NANPDB2476 had binding affinities of −8.2, −8.2, −8.1 and −8.0 kcal/mol, respectively, and were predicted to possess desirable antiviral activity including the inhibition of RNA synthesis and membrane permeability, with the probable activity (Pa) being greater than the probable inactivity (Pi) values. The predicted anti-EBOV inhibition efficiency values (IC50), found using a random forest classifier, ranged from 3.35 to 11.99 μM, while the Ki values ranged from 0.97 to 1.37 μM. The compounds NANPDB4048 and NANPDB2412 had the lowest binding energy of −8.2 kcal/mol, implying a higher binding affinity to EBOV VP35 which was greater than those of the known inhibitors. The compounds were predicted to possess a low toxicity risk and to possess reasonably good pharmacological profiles. Molecular dynamics (MD) simulations of the protein–ligand complexes, lasting 50 ns, and molecular mechanisms Poisson-Boltzmann surface area (MM-PBSA) calculations corroborated the binding affinities of the identified compounds and identified novel critical interacting residues. The antiviral potential of the molecules could be confirmed experimentally, while the scaffolds could be optimized for the design of future novel anti-EBOV chemotherapeutics.

show abstract

Section: Methodsmentioning

confidence: 99%

Computational Study on Potential Novel Anti-Ebola Virus Protein VP35 Natural Compounds

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In addition, they can predict the impact of materials not yet synthesized, thereby contributing to safe-by-design approaches [ 51 ]. ML has been effectively employed for the prediction of toxicity profiles of NPs [ 52 , 53 , 54 , 55 , 56 ] and for the development of new antibiotics [ 57 , 58 ]. Furthermore, models for the prediction of the antimicrobial resistance for specific bacteria have been demonstrated [ 59 , 60 , 61 ].…”

Section: Introductionmentioning

confidence: 99%

A Machine Learning Tool to Predict the Antibacterial Capacity of Nanoparticles

et al. 2021

View full text Add to dashboard Cite

The emergence and rapid spread of multidrug-resistant bacteria strains are a public health concern. This emergence is caused by the overuse and misuse of antibiotics leading to the evolution of antibiotic-resistant strains. Nanoparticles (NPs) are objects with all three external dimensions in the nanoscale that varies from 1 to 100 nm. Research on NPs with enhanced antimicrobial activity as alternatives to antibiotics has grown due to the increased incidence of nosocomial and community acquired infections caused by pathogens. Machine learning (ML) tools have been used in the field of nanoinformatics with promising results. As a consequence of evident achievements on a wide range of predictive tasks, ML techniques are attracting significant interest across a variety of stakeholders. In this article, we present an ML tool that successfully predicts the antibacterial capacity of NPs while the model’s validation demonstrates encouraging results (R2 = 0.78). The data were compiled after a literature review of 60 articles and consist of key physico-chemical (p-chem) properties and experimental conditions (exposure variables and bacterial clustering) from in vitro studies. Following data homogenization and pre-processing, we trained various regression algorithms and we validated them using diverse performance metrics. Finally, an important attribute evaluation, which ranks the attributes that are most important in predicting the outcome, was performed. The attribute importance revealed that NP core size, the exposure dose, and the species of bacterium are key variables in predicting the antibacterial effect of NPs. This tool assists various stakeholders and scientists in predicting the antibacterial effects of NPs based on their p-chem properties and diverse exposure settings. This concept also aids the safe-by-design paradigm by incorporating functionality tools.

show abstract

“…However the large amount of data available promotes the use of machine learning techniques in discovery projects (for example, building regression, classification models and virtual classification or selection of compounds). The authors of [77] review some Machine Learning applications focusing on the development of new antibiotics, the prediction of resistance and its mechanisms.…”

Section: Biological Problems Asses By Machine Learning In Drug Discoverymentioning

confidence: 99%

A review on machine learning approaches and trends in drug discovery

Carracedo-Reboredo

Liñares-Blanco

Rodríguez-Fernández

et al. 2021

Computational and Structural Biotechnology Journal

268

120

View full text Add to dashboard Cite

show abstract

The application of machine learning techniques to innovative antibacterial discovery and development

Cited by 37 publications

References 114 publications

Computational Study on Potential Novel Anti-Ebola Virus Protein VP35 Natural Compounds

Computational Study on Potential Novel Anti-Ebola Virus Protein VP35 Natural Compounds

A Machine Learning Tool to Predict the Antibacterial Capacity of Nanoparticles

A review on machine learning approaches and trends in drug discovery

Contact Info

Product

Resources

About