Computational QSAR model combined molecular descriptors and fingerprints to predict HDAC1 inhibitors

Stack-HDAC3i: A high-precision identification of HDAC3 inhibitors by exploiting a stacked ensemble-learning framework

Epigenetics involves reversible modifications in gene expression without altering the genetic code itself. Among these modifications, histone deacetylases (HDACs) play a key role by removing acetyl groups from lysine residues on histones. Overexpression of HDACs is linked to the proliferation and survival of tumor cells. To combat this, HDAC inhibitors (HDACi) are commonly used in cancer treatments. However, pan-HDAC inhibition can lead to numerous side effects. Therefore, isoform-selective HDAC inhibitors, such as HDAC3i, could be advantageous for treating various medical conditions while minimizing off-target effects. To date, computational approaches that use only the SMILES notation without any experimental evidence have become increasingly popular and necessary for the initial discovery of novel potential therapeutic drugs. In this study, we develop an innovative and high-precision stacked-ensemble framework, called Stack-HDAC3i, which can directly identify HDAC3i using only the SMILES notation. Using an up-to-date benchmark dataset, we first employed both molecular descriptors and Mol2Vec embeddings to generate feature representations that cover multi-view information embedded in HDAC3i, such as structural and contextual information. Subsequently, these feature representations were used to train baseline models using nine popular ML algorithms. Finally, the probabilistic features derived from the selected baseline models were fused to construct the final stacked model. Both cross-validation and independent tests showed that Stack-HDAC3i is a high-accuracy prediction model with great generalization ability for identifying HDAC3i. Furthermore, in the independent test, Stack-HDAC3i achieved an accuracy of 0.926 and Matthew's correlation coefficient of 0.850, which are 0.44-6.11% and 0.83-11.90% higher than its constituent baseline models, respectively.

Descriptor generation from Morgan fingerprint using persistent homology

In cheminformatics, molecular fingerprints (FPs) are used in various tasks such as regression and classification. However, predictive models often underutilize Morgan FP for regression and related tasks in machine learning. This study introduced descriptors derived from reshaped Morgan FPs using persistent homology for the predictive accuracy improvement. In the solvation free energy (FreeSolv) and water solubility (ESOL) datasets, persistent homology was found to enhance predictive accuracy compared to the use of only Morgan FPs. Notably, using the first-order persistence diagram (PD1) for descriptor generation resulted in more significant improvements than using the zeroth-order persistence diagram (PD0). Combining 4096 bits Morgan FPs with PD1-generated descriptors increased the average coefficient of determination in the Gaussian process regression from 0.597 to 0.667 for FreeSolv and from 0.629 to 0.654 for ESOL. Adjusting the grid size parameter during PD-based descriptor generation is crucial, as finer grids, especially with PD0, generate more descriptors but reduce predictive accuracy. Coarsening the grid or applying principal component analysis (PCA) mitigates overfitting and enhances accuracy. When descriptors were generated from Morgan FPs with randomly shuffled bit positions, coarsening the grid and/or applying PCA achieved similar accuracy improvements as when the persistent homology of the original Morgan FPs was used.

Virtual screening and invitro evaluation of cyclooxygenase inhibitors from Tinospora cordifolia using the machine learning tool

Tinospora cordifolia has a variety of compounds, and some of these compounds may have anti-inflammatory and antioxidant properties. In the present study, we identified the compounds in the leaf extract of T. cordifolia through Gas Chromatography-Mass Spectrometry (GC-MS) analysis and found the various metabolites. The compounds are screened virtually using a machine learning model, followed by molecular docking and simulation study to identify top-hit compounds as cyclooxygenase (COX) inhibitors. The molecular docking revealed that the compound 7,9-Di-tert-butyl-1-oxaspiro (4,5) deca-6,9-diene-2,8-dione (CID:545303) exhibited the lowest binding energies of -7.1 and -6.8 kcal/mol against COX 1 and COX 2 respectively. The interactions are favored by hydrogen bonding and hydrophobic interaction inside the binding pocket. The 100 ns MD simulation study for these compounds was performed to know the stability and found the RMSD around 2 Å and around 1.0 Å with minimal fluctuations indicating a stable complex throughout the simulation of 100 ns. Based on these findings, we proposed 7,9-Di-tertbutyl- 1-oxaspiro (4,5) deca-6,9-diene-2,8-dione could be used as a dual inhibitor of COX enzymes and a drug-like molecule for treating inflammation after evaluation of their biological properties. The methanolic extract of T. cordifolia was subjected to in vitro DPPH, ABTS, nitric oxide, anti-microbial, COX, and LOX inhibition activity. The results exhibited possible positive effects against the above activities.Communicated by Ramaswamy H. Sarma.

HDAC1 PREDICTOR: a simple and transparent application for virtual screening of histone deacetylase 1 inhibitors

Histone deacetylases play an important role in regulating gene expression by modifying histones and changing chromatin conformation. HDAC dysregulation is involved in many diseases, such as cancer, autoimmune and neurodegenerative diseases. Histone deacetylase 1 (HDAC1) inhibitors represent an important class of drugs. Quantitative Structure-Activity Relationship (QSAR) classification models were developed using 2D RDKit molecular descriptors; ECPF4 (Extended Connectivity Fingerprint) circular fingerprints; and the Random Forest, Gradient Boosting, and Support Vector Machine methods. The developed models were integrated into the HDAC1 PREDICTOR application, which is freely available at the link https://ovttiras-hdac1-inhibitors-hdac1-predictor-app-z3mrbr.streamlitapp.com. The HDAC1 PREDICTOR web application allows one to reveal the compounds for which the predicted activity to inhibit HDAC1 is higher than that of the reference Vorinostat compound (IC₅₀ = 11.08 nM). The algorithm implemented in HDAC1 PREDICTOR for determining the contributions of molecular fragments to the inhibitory activity can be used to find the molecule segments that increase or decrease the activity, enabling the researcher to conduct a rational molecular design of new highly active HDAC1 inhibitors. The developed QSAR models and the code for their construction in the Python programming language are freely available on the GitHub platform at https://github.com/ovttiras/HDAC1-inhibitors.

A quantitative structural analysis of AR-42 derivatives as HDAC1 inhibitors for the identification of promising structural contributors

Alteration and abnormal epigenetic mechanisms can lead to the aberration of normal biological functions and the occurrence of several diseases. The histone deacetylase (HDAC) family of enzymes is one of the prime regulators of epigenetic functions modifying the histone proteins, and thus, regulating epigenetics directly. HDAC1 is one of those HDACs which have important contributions to cellular epigenetics. The abnormality of HDAC is correlated to the occurrence, progression, and poor prognosis in several disease conditions namely neurodegenerative disorders, cancer cell proliferation, metastasis, chemotherapy resistance, and survival in various cancers. Therefore, the progress of potent and effective HDAC1 inhibitors is one of the prime approaches to combat such diseases. In this study, both regression and classification-based molecular modelling studies were conducted on some AR-42 derivatives as HDAC1 inhibitors to elucidate the crucial structural aspects that are responsible for regulating their biological responses. This study revealed that the molecular polarizability, van der Waals volume, the presence of aromatic rings as well as the higher number of hydrogen bond acceptors might affect prominently their inhibitory activity and might be responsible for proper fitting and interactions at the HDAC1 active site to pertain effective inhibition.

Identification of novel leads as potent inhibitors of HDAC3 using ligand-based pharmacophore modeling and MD simulation

In the landscape of epigenetic regulation, histone deacetylase 3 (HDAC3) has emerged as a prominent therapeutic target for the design and development of candidate drugs against various types of cancers and other human disorders. Herein, we have performed ligand-based pharmacophore modeling, virtual screening, molecular docking, and MD simulations to design potent and selective inhibitors against HDAC3. The predicted best pharmacophore model 'Hypo 1' showed excellent correlation (R2 = 0.994), lowest RMSD (0.373), lowest total cost value (102.519), and highest cost difference (124.08). Hypo 1 consists of four salient pharmacophore features viz. one hydrogen bond acceptor (HBA), one ring aromatic (RA), and two hydrophobic (HYP). Hypo 1 was validated by Fischer's randomization with a 95% of confidence level and the external test set of 60 compounds with a good correlation coefficient (R2 = 0.970). The virtual screening of chemical databases, drug-like properties calculations followed by molecular docking resulted in identifying 22 representative hit compounds. Performed 50 ns of MD simulations on top three hits were retained the salient π-stacking, Zn2+ coordination, hydrogen bonding, and hydrophobic interactions with catalytic residues from the active site pocket of HDAC3. Total binding energy calculated by MM-PBSA showed that the Hit 1 and Hit 2 formed stable complexes with HDAC3 as compared to reference TSA. Further, the PLIP analysis showed a close resemblance between the salient pharmacophore features of Hypo 1 and the presence of molecular interactions in co-crystallized FDA-approved drugs. We conclude that the screened hit compounds may act as potent inhibitors of HDAC3 and further preclinical and clinical studies may pave the way for developing them as effective therapeutic agents for the treatment of different cancers and neurodegenerative disorders.

Virtual screening of dipeptidyl peptidase-4 inhibitors using quantitative structure-activity relationship-based artificial intelligence and molecular docking of hit compounds

Dipeptidyl peptidase-4 (DPP-4) inhibitors are becoming an essential drug in the treatment of type 2 diabetes mellitus; however, some classes of these drugs exert side effects, including joint pain and pancreatitis. Studies suggest that these side effects might be related to secondary inhibition of DPP-8 and DPP-9. In this study, we identified DPP-4-inhibitor hit compounds selective against DPP-8 and DPP-9. We built a virtual screening workflow using a quantitative structure-activity relationship (QSAR) strategy based on artificial intelligence to allow faster screening of millions of molecules for the DPP-4 target relative to other screening methods. Five regression machine learning algorithms and four classification machine learning algorithms were applied to build virtual screening workflows, with the QSAR model applied using support vector regression (R²_pred 0.78) and the classification QSAR model using the random forest algorithm with 92.2% accuracy. Virtual screening results of > 10 million molecules obtained 2 716 hits compounds with a pIC₅₀ value of > 7.5. Additionally, molecular docking results of several potential hit compounds for DPP-4, DPP-8, and DPP-9 identified CH0002 as showing high inhibitory potential against DPP-4 and low inhibitory potential for DPP-8 and DPP-9 enzymes. These results demonstrated the effectiveness of this technique for identifying DPP-4-inhibitor hit compounds selective for DPP-4 and against DPP-8 and DPP-9 and suggest its potential efficacy for applications to discover hit compounds of other targets.

Artificial Intelligence for Drug Toxicity and Safety

Interventional pharmacology is one of medicine's most potent weapons against disease. These drugs, however, can result in damaging side effects and must be closely monitored. Pharmacovigilance is the field of science that monitors, detects, and prevents adverse drug reactions (ADRs). Safety efforts begin during the development process, using in vivo and in vitro studies, continue through clinical trials, and extend to postmarketing surveillance of ADRs in real-world populations. Future toxicity and safety challenges, including increased polypharmacy and patient diversity, stress the limits of these traditional tools. Massive amounts of newly available data present an opportunity for using artificial intelligence (AI) and machine learning to improve drug safety science. Here, we explore recent advances as applied to preclinical drug safety and postmarketing surveillance with a specific focus on machine and deep learning (DL) approaches.

Small molecule HDAC inhibitors: Promising agents for breast cancer treatment

Breast cancer, a heterogeneous disease, is the most frequently diagnosed cancer and the second leading cause of cancer-related death among women worldwide. Recently, epigenetic abnormalities have emerged as an important hallmark of cancer development and progression. Given that histone deacetylases (HDACs) are crucial to chromatin remodeling and epigenetics, their inhibitors have become promising potential anticancer drugs for research. Here we reviewed the mechanism and classification of histone deacetylases (HDACs), association between HDACs and breast cancer, classification and structure-activity relationship (SAR) of HDACIs, pharmacokinetic and toxicological properties of the HDACIs, and registered clinical studies for breast cancer treatment. In conclusion, HDACIs have shown desirable effects on breast cancer, especially when they are used in combination with other anticancer agents. In the coming future, more multicenter and randomized Phase III studies are expected to be conducted pushing promising new therapies closer to the market. In addition, the design and synthesis of novel HDACIs are also needed.

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.