Artificial intelligence and machine-learning approaches in structure and ligand-based discovery of drugs affecting central nervous system.

CNS disorders are indications with a very high unmet medical needs, relatively smaller number of available drugs, and a subpar satisfaction level among patients and caregiver. Discovery of CNS drugs is extremely expensive affair with its own unique challenges leading to extremely high attrition rates and low efficiency. With explosion of data in information age, there is hardly any aspect of life that has not been touched by data driven technologies such as artificial intelligence (AI) and machine learning (ML). Drug discovery is no exception, emergence of big data via genomic, proteomic, biological, and chemical technologies has driven pharmaceutical giants to collaborate with AI oriented companies to revolutionise drug discovery, with the goal of increasing the efficiency of the process. In recent years many examples of innovative applications of AI and ML techniques in CNS drug discovery has been reported. Research on therapeutics for diseases such as schizophrenia, Alzheimer's and Parkinsonism has been provided with a new direction and thrust from these developments. AI and ML has been applied to both ligand-based and structure-based drug discovery and design of CNS therapeutics. In this review, we have summarised the general aspects of AI and ML from the perspective of drug discovery followed by a comprehensive coverage of the recent developments in the applications of AI/ML techniques in CNS drug discovery.

Computational Approaches in the Discovery and Development of Therapeutic and Prophylactic Agents for Viral Diseases.

Over the last two decades, computational technologies have played a crucial role in antiviral drug development. Whenever a virus spreads and becomes a threat to global health, it brings along the challenge of developing new therapeutics and prophylactics. Computational drug and vaccine discovery has evolved quickly over the years. Some interesting examples of computational drug discovery are anti-AIDS drugs, where HIV protease and reverse transcriptase have been targeted by agents developed using computational methods. Various computational methods that have been applied to anti-viral research include ligand-based methods that rely on known active compounds, i.e., pharmacophore modeling, machine learning or classical QSAR; structure-based methods that rely on an experimentally determined 3D structure of the targets, i.e., molecular docking and molecular dynamics and methods for the development of vaccines such as reverse vaccinology; structural vaccinology and vaccine epitope prediction. This review summarizes these approaches to battle viral diseases and underscores their importance for anti-viral research. We discuss the role of computational methods in developing small molecules and vaccines against human immunodeficiency virus, yellow fever, human papilloma virus, SARS-CoV-2, and other viruses. Various computational tools available for the abovementioned purposes have been listed and described. A discussion on applying artificial intelligence-based methods for antiviral drug discovery has also been included.

Molecular Dynamics Simulations in Designing DARPins as Phosphorylation-Specific Protein Binders of ERK2.

Extracellular signal-regulated kinases 1 and 2 (ERK1/2) play key roles in promoting cell survival and proliferation through the phosphorylation of various substrates. Remarkable antitumour activity is found in many inhibitors that act upstream of the ERK pathway. However, drug-resistant tumour cells invariably emerge after their use due to the reactivation of ERK1/2 signalling. ERK1/2 inhibitors have shown clinical efficacy as a therapeutic strategy for the treatment of tumours with mitogen-activated protein kinase (MAPK) upstream target mutations. These inhibitors may be used as a possible strategy to overcome acquired resistance to MAPK inhibitors. Here, we report a class of repeat proteins-designed ankyrin repeat protein (DARPin) macromolecules targeting ERK2 as inhibitors. The structural basis of ERK2-DARPin interactions based on molecular dynamics (MD) simulations was studied. The information was then used to predict stabilizing mutations employing a web-based algorithm, MAESTRO. To evaluate whether these design strategies were successfully deployed, we performed all-atom, explicit-solvent molecular dynamics (MD) simulations. Two mutations, Ala → Asp and Ser → Leu, were found to perform better than the original sequence (DARPin E40) based on the associated energy and key residues involved in protein-protein interaction. MD simulations and analysis of the data obtained on these mutations supported our predictions.

Pharmacophore Based Virtual Screening Approach to Identify Selective PDE4B Inhibitors.

Phosphodiesterase 4 (PDE4) has been established as a promising target in asthma and chronic obstructive pulmonary disease. PDE4B subtype selective inhibitors are known to reduce the dose limiting adverse effect associated with non-selective PDE4B inhibitors. This makes the development of PDE4B subtype selective inhibitors a desirable research goal. To achieve this goal, ligand based pharmacophore modeling approach is employed. Separate pharmacophore hypotheses for PDE4B and PDE4D inhibitors were generated using HypoGen algorithm and 106 PDE4 inhibitors from literature having thiopyrano [3,2-d] Pyrimidines, 2-arylpyrimidines, and triazines skeleton. Suitable training and test sets were created using the molecules as per the guidelines available for HypoGen program. Training set was used for hypothesis development while test set was used for validation purpose. Fisher validation was also used to test the significance of the developed hypothesis. The validated pharmacophore hypotheses for PDE4B and PDE4D inhibitors were used in sequential virtual screening of zinc database of drug like molecules to identify selective PDE4B inhibitors. The hits were screened for their estimated activity and fit value. The top hit was subjected to docking into the active sites of PDE4B and PDE4D to confirm its selectivity for PDE4B. The hits are proposed to be evaluated further using in-vitro assays.

Structure-based design of selective phosphodiesterase 4B inhibitors based on ginger phenolic compounds.

Phosphodiesterase 4 (PDE4) has been established as a drug target for inflammatory diseases of respiratory tract like asthma and chronic obstructive pulmonary disease. The selective inhibitors of PDE4B, a subtype of PDE4, are devoid of adverse effects like nausea and vomiting commonly associated with non-selective PDE4B inhibitors. This makes the development of PDE4B subtype selective inhibitors a desirable research goal. Thus, in the present study, molecular docking, molecular dynamic simulations and binding free energy were performed to explore potential selective PDE4B inhibitors based on ginger phenolic compounds. The results of docking studies indicate that some of the ginger phenolic compounds demonstrate higher selective PDE4B inhibition than existing selective PDE4B inhibitors. Additionally, 6-gingerol showed the highest PDE4B inhibitory activity as well as selectivity. The comparison of binding mode of PDE4B/6-gingerol and PDE4D/6-gingerol complexes revealed that 6-gingerol formed additional hydrogen bond and hydrophobic interactions with active site and control region 3 (CR3) residues in PDE4B, which were primarily responsible for its PDE4B selectivity. The results of binding free energy demonstrated that electrostatic energy is the primary factor in elucidating the mechanism of PDE4B inhibition by 6-gingerol. Dynamic cross-correlation studies also supported the results of docking and molecular dynamics simulation. Finally, a small library of molecules were designed based on the identified structural features, majority of designed molecules showed higher PDE4B selectivity than 6-gingerol. These results provide important structural features for designing new selective PDE4B inhibitors as anti-inflammatory drugs and promising candidates for synthesis and pre-clinical pharmacological investigations.

QSAR studies on 4-quinolone derivatives as high-affinity ligands at the benzodiazepine site of brain GABA(A) receptors.

Quantitative structure activity relationship approach using stepwise regression analysis was applied to a series of 4-quinolone derivatives as high-affinity ligands at the benzodiazepine site of brain GABA(A) receptors. For the purpose 25 compounds were used to develop models. Statistically significant equations were obtained with high squared correlation coefficient (r(2)=0.8761, 0.9295 and 0.8705) and low root mean square error (RMSE=0.4844, 0.3894 and 0.4952). The robustness of the model was confirmed with the help of leave one out cross validation method which exhibited high r(2)(cv) values (r(2)(cv)=0.7875, 0.8263 and 0.7927). A good correlation of various molecular shape parameters, like ovality, Szeged index, and energy of the molecule with the GABA(A) affinity was achieved.