Natural product drug discovery in the artificial intelligence era.

Natural products (NPs) are primarily recognized as privileged structures to interact with protein drug targets. Their unique characteristics and structural diversity continue to marvel scientists for developing NP-inspired medicines, even though the pharmaceutical industry has largely given up. High-performance computer hardware, extensive storage, accessible software and affordable online education have democratized the use of artificial intelligence (AI) in many sectors and research areas. The last decades have introduced natural language processing and machine learning algorithms, two subfields of AI, to tackle NP drug discovery challenges and open up opportunities. In this article, we review and discuss the rational applications of AI approaches developed to assist in discovering bioactive NPs and capturing the molecular "patterns" of these privileged structures for combinatorial design or target selectivity.

Synthesis of covalent bonding MWCNT-oligoethylene linezolid conjugates and their antibacterial activity against bacterial strains.

Nowadays, infectious diseases caused by drug-resistant bacteria have become especially important. Linezolid is an antibacterial drug active against clinically important Gram positive strains; however, resistance showed by these bacteria has been reported. Nanotechnology has improved a broad area of science, such as medicine, developing new drug delivery and transport systems. In this work, several covalently bounded conjugated nanomaterials were synthesized from multiwalled carbon nanotubes (MWCNTs), a different length oligoethylene chain (S n ), and two linezolid precursors (4 and 7), and they were evaluated in antibacterial assays. Interestingly, due to the intrinsic antibacterial activity of the amino-oligoethylene linezolid analogues, these conjugated nanomaterials showed significant antibacterial activity against various tested bacterial strains in a radial diffusion assay and microdilution method, including Gram negative strains as Escherichia coli (11 mm, 6.25 µg mL-1) and Salmonella typhi (14 mm, ≤0.78 µg mL-1), which are not inhibited by linezolid. The results show a significant effect of the oligoethylene chain length over the antibacterial activity. Molecular docking of amino-oligoethylene linezolid analogs shows a more favorable interaction of the S 2-7 analog in the PTC of E. coli.

Conformal prediction of HDAC inhibitors.

The growing interest in epigenetic probes and drug discovery, as revealed by several epigenetic drugs in clinical use or in the lineup of the drug development pipeline, is boosting the generation of screening data. In order to maximize the use of structure-activity relationships there is a clear need to develop robust and accurate models to understand the underlying structure-activity relationship. Similarly, accurate models should be able to guide the rational screening of compound libraries. Herein we introduce a novel approach for epigenetic quantitative structure-activity relationship (QSAR) modelling using conformal prediction. As a case study, we discuss the development of models for 11 sets of inhibitors of histone deacetylases (HDACs), which are one of the major epigenetic target families that have been screened. It was found that all derived models, for every HDAC endpoint and all three significance levels, are valid with respect to predictions for the external test sets as well as the internal validation of the corresponding training sets. Furthermore, the efficiencies for the predictions are above 80% for most data sets and above 90% for four data sets at different significant levels. The findings of this work encourage prospective applications of conformal prediction for other epigenetic target data sets.

Conformation-dependent QSAR approach for the prediction of inhibitory activity of bromodomain modulators.

Epigenetic drug discovery is a promising research field with growing interest in the scientific community, as evidenced by the number of publications and the large amount of structure-epigenetic activity information currently available in the public domain. Computational methods are valuable tools to analyse and understand the activity of large compound collections from their structural information. In this manuscript, QSAR models to predict the inhibitory activity of a diverse and heterogeneous set of 88 organic molecules against the bromodomains BRD2, BRD3 and BRD4 are presented. A conformation-dependent representation of the chemical structures was established using the RDKit software and a training and test set division was performed. Several two-linear and three-linear QuBiLS-MIDAS molecular descriptors ( www.tomocomd.com ) were computed to extract the geometric structural features of the compounds studied. QuBiLS-MIDAS-based features sets, to be used in the modelling, were selected using dimensionality reduction strategies. The multiple linear regression procedure coupled with a genetic algorithm were employed to build the predictive models. Regression models containing between 6 to 9 variables were developed and assessed according to several internal and external validation methods. Analyses of outlier compounds and the applicability domain for each model were performed. As a result, the models against BRD2 and BRD3 with 8 variables and the model with 9 variables against BRD4 were those with the best overall performance according to the criteria accounted for. The results obtained suggest that the models proposed will be a good tool for studying the inhibitory activities of drug candidates against the bromodomains considered during epigenetic drug discovery.

Molecular Modeling and Chemoinformatics to Advance the Development of Modulators of Epigenetic Targets: A Focus on DNA Methyltransferases.

In light of the emerging field of Epi-informatics, ie, computational methods applied to epigenetic research, molecular docking, and dynamics, pharmacophore and activity landscape modeling and QSAR play a key role in the development of modulators of DNA methyltransferases (DNMTs), one of the major epigenetic target families. The increased chemical information available for modulators of DNMTs has opened up the avenue to explore the epigenetic relevant chemical space (ERCS). Herein, we discuss recent progress on the identification and development of inhibitors of DNMTs as potential epi-drugs and epi-probes that have been driven by molecular modeling and chemoinformatics methods. We also survey advances on the elucidation of their structure-activity relationships and exploration of ERCS. Finally, it is illustrated how computational approaches can be applied to identify modulators of DNMTs in food chemicals.

Inhibitors of DNA methyltransferases: insights from computational studies.

DNA methyltransferases (DNMTs) are a family of epigenetic enzymes for which inhibition is an attractive strategy for the treatment of cancer and other diseases. In synergy with experimental approaches, computational methods are increasingly being used to identify and optimize the activity of inhibitors of DNMTs as well as to rationalize at the molecular level of the mechanism of established inhibitors. Recently, a crystallographic structure of the methyltransferase domain of human DNMT1 bound to unmethylated DNA was published encouraging the application of structure-based approaches to design and optimize the activity of currently known inhibitors. Herein, we review the progress in the discovery and optimization of inhibitors of DNMTs using computational approaches including homology modeling, docking, pharmacophore modeling, molecular dynamics, and virtual screening.

How to recognize and workaround pitfalls in QSAR studies: a critical review.

Quantitative Structure-Activity Relationships (QSAR) are based on the hypothesis that changes in molecular structure reflect proportional changes in the observed response or biological activity. In order to successfully conduct QSAR studies certain conditions have to be met that are not frequently reported in the literature. This suggests that some authors are not aware of the principle flaws, occasional shortcomings, and circumstantial downsides of QSAR methods. The present paper focuses on prerequisites to set up correct models and on limitations of model applications. Their implications are systematically described and illustrated as pitfalls that have strong implications in QSAR, and possible solutions are suggested. The paper is focused on small scale 2D- and 3D-QSAR studies for lead optimization. The work is enriched with comprehensive comments and non-mathematical explanations for the computer practitioner in Medicinal Chemistry.

Docking-based CoMFA and CoMSIA studies of non-nucleoside reverse transcriptase inhibitors of the pyridinone derivative type.

Comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) were performed on a set of pyridinone derivatives. A molecular alignment obtained by docking of compounds into the non-nucleoside reverse transcriptase inhibitor binding site of HIV-1 was used. Good correlations between the calculated binding free energies and experimental inhibitory activities suggest that the binding conformations of these inhibitors are reasonable. Robust and predictive 3D-QSAR models were obtained with q2 values of 0.706 and 0.723 for CoMFA and CoMSIA, respectively. The models were validated by an external test set obtaining r2 pred values of 0.720 and 0.750 for CoMFA and CoMSIA, respectively. The CoMFA, CoMSIA and docking results help to understand the type of interactions that occur between pyridinone derivatives with the non-nucleoside reverse transcriptase inhibitor binding pocket, and explain the viral resistance to pyridinone derivatives upon mutation of amino acids Tyr181 and Tyr188. The results obtained provide information for a better understanding of the drug resistant mechanisms. The 3D-QSAR models derived will be used to guide the design of pyridinone derivatives active against mutant strains of reverse transcriptase.