PyPLIF HIPPOS: A Molecular Interaction Fingerprinting Tool for Docking Results of AutoDock Vina and PLANTS

PredCoffee: A binary classification approach specifically for coffee odor

Compared to traditional methods, using machine learning to assess or predict the odor of molecules can save costs in various aspects. Our research aims to collect molecules with coffee odor and summarize the regularity of these molecules, ultimately creating a binary classifier that can determine whether a molecule has a coffee odor. In this study, a total of 371 coffee-odor molecules and 9,700 non-coffee-odor molecules were collected. The Knowledge-guided Pre-training of Graph Transformer (KPGT), support vector machine (SVM), random forest (RF), multi-layer perceptron (MLP), and message-passing neural networks (MPNN) were used to train the data. The model with the best performance was selected as the basis of the predictor. The prediction accuracy value of the KPGT model exceeded 0.84 and the predictor has been deployed as a webserver PredCoffee.

Baicalein attenuates oxidative damage in mice haematopoietic cells through regulation of PDGFRβ

Platelet-derived growth factor receptor β (PDGFRβ) plays a crucial role in murine haematopoiesis. Baicalein (BAI), a naturally occurring flavonoid, can alleviate disease damage through anti-oxidative, anti-apoptotic, and anti-inflammatory mechanisms. However, whether BAI attenuates oxidative damage in murine haematopoietic cells by PDGFRβ remains unexplored. In this study, we utilized a tert-butyl hydroperoxide (TBHP)-induced BaF3 cell injury model and an ionising radiation (IR)-induced mice injury model to investigate the impact of the presence or absence of PDGFRβ on the pharmacological effects of BAI. In addition, the BAI-PDGFRβ interaction was characterized by molecular docking and dynamics simulations. The results show that a specific concentration of BAI led to increased cell viability, reduced reactive oxygen species (ROS) content, upregulated nuclear factor erythroid 2-related factor 2 (NRF2) expression, and its downstream target genes heme oxygenase 1 (HO-1) and NAD(P)H Quinone Dehydrogenase 1 (NQO1), and activated protein kinase B (AKT) pathway in cells expressing PDGFRβ plasmid and experiencing damage. Similarly, BAI elevated lineage-Sca1+cKIT+ (LSK) cell proportion, promoted haematopoietic restoration, enhanced NRF2-mediated antioxidant response in PDGFRβ+/+ mice. However, despite BAI usage, PDGFRβ knockout mice (PDGFRβ-/-) showed lower LSK proportion and less antioxidant capacity than the total body irradiation (TBI) group. Furthermore, we demonstrated an interaction between BAI and PDGFRβ at the molecular level. Collectively, our results indicate that BAI attenuates oxidative stress injury and helps promote haematopoietic cell recovery through regulation of PDGFRβ.

Hypoglycemic Activity of Rice Resistant-Starch Metabolites: A Mechanistic Network Pharmacology and In Vitro Approach

Rice (Oryza sativa L.) is one of the primary sources of energy and nutrients needed by the body, and rice resistant starch (RRS) has been found to have hypoglycemic effects. However, its biological activity and specific mechanisms still need to be further elucidated. In the present study, 52 RRS differential metabolites were obtained from mouse liver, rat serum, canine feces, and human urine, and 246 potential targets were identified through a literature review and database analysis. A total of 151 common targets were identified by intersecting them with the targets of type 2 diabetes mellitus (T2DM). After network pharmacology analysis, 11 core metabolites were identified, including linolenic acid, chenodeoxycholic acid, ursodeoxycholic acid, deoxycholic acid, lithocholic acid, lithocholylglycine, glycoursodeoxycholic acid, phenylalanine, norepinephrine, cholic acid, and L-glutamic acid, and 16 core targets were identified, including MAPK3, MAPK1, EGFR, ESR1, PRKCA, FYN, LCK, DLG4, ITGB1, IL6, PTPN11, RARA, NR3C1, PTPN6, PPARA, and ITGAV. The core pathways included the neuroactive ligand-receptor interaction, cancer, and arachidonic acid metabolism pathways. The molecular docking results showed that bile acids such as glycoursodeoxycholic acid, chenodeoxycholic acid, ursodeoxycholic acid, lithocholic acid, deoxycholic acid, and cholic acid exhibited strong docking effects with EGFR, ITGAV, ITGB1, MAPK3, NR3C1, α-glucosidase, and α-amylase. In vitro hypoglycemic experiments further suggested that bile acids showed significant inhibitory effects on α-glucosidase and α-amylase, with CDCA and UDCA having the most prominent inhibitory effect. In summary, this study reveals a possible hypoglycemic pathway of RRS metabolites and provides new research perspectives to further explore the therapeutic mechanism of bile acids in T2DM.

Building a Kokumi Database and Machine Learning-Based Prediction: A Systematic Computational Study on Kokumi Analysis

Kokumi is a subtle sensation characterized by a sense of fullness, continuity, and thickness. Traditional methods of taste discovery and analysis, including those of kokumi, have been labor-intensive and costly, thus necessitating the emergence of computational methods as critical strategies in molecular taste analysis and prediction. In this study, we undertook a comprehensive analysis, prediction, and screening of the kokumi compounds. We categorized 285 kokumi compounds from a previously unreleased kokumi database into five groups based on their molecular characteristics. Moreover, we predicted kokumi/non-kokumi and multi-flavor compositions using six structure-taste relationship models: MLP-E3FP, MLP-PLIF, MLP-RDKFP, SVM-RDKFP, RF-RDKFP, and WeaveGNN feature of Atoms and Bonds. These six predictors exhibited diverse performance levels across two different models. For kokumi/non-kokumi prediction, the WeaveGNN model showed an exceptional predictive AUC value (0.94), outperforming the other models (0.87, 0.90, 0.89, 0.92, and 0.78). For multi-flavor prediction, the MLP-E3FP model demonstrated a higher predictive AUC and MCC value (0.94 and 0.74) than the others (0.73 and 0.33; 0.92 and 0.70; 0.95 and 0.73; 0.94 and 0.64; and 0.88 and 0.69). This data highlights the model's proficiency in accurately predicting kokumi molecules. As a result, we sourced kokumi active compounds through a high-throughput screening of over 100 million molecules, further refined by toxicity and similarity screening. Lastly, we launched a web platform, KokumiPD (https://www.kokumipd.com/), offering a comprehensive kokumi database and online prediction services for users.

In-silico prediction of anti-breast cancer activity of ginger (Zingiber officinale) using machine learning techniques

INTRODUCTION

Indonesian civilization extensively uses traditional medicine to cure illnesses and preserve health. The lack of knowledge on the security and efficacy of medicinal plants is still a significant concern. Although the precise chemicals responsible for this impact are unknown, ginger is a common medicinal plant in Southeast Asia that may have anticancer qualities.

METHOD

Using data from Dudedocking, a machine-learning model was created to predict possible breast anticancer chemicals from ginger. The model was used to forecast substances that block KIT and MAPK2 proteins, essential elements in breast cancer.

RESULT

Beta-carotene, 5-Hydroxy-74'-dimethoxyflavone, [12]-Shogaol, Isogingerenone B, curcumin, Trans-[10]-Shogaol, Gingerenone A, Dihydrocurcumin, and demethoxycurcumin were all superior to the reference ligand for MAPK2, according to molecular docking studies. Lycopene, [8]-Shogaol, [6]-Shogaol, and [1]-Paradol exhibited low toxicity and no Lipinski violations, but beta carotene had toxic predictions and Lipinski violations. It was anticipated that all three substances would have anticarcinogenic qualities.

CONCLUSION

Overall, this study shows the value of machine learning in drug development and offers insightful information on possible anticancer chemicals from ginger.

Caffeic Acid in Spent Coffee Grounds as a Dual Inhibitor for MMP-9 and DPP-4 Enzymes

Type 2 diabetes mellitus and diabetic foot ulcers remain serious worldwide health problems. Caffeic acid is one of the natural products that has been experimentally proven to have diverse pharmacological properties. This study aimed to assess the inhibitory activity of caffeic acid and ethanolic extract of spent coffee grounds targeting DPP-4 and MMP-9 enzymes and evaluate the molecular interactions through 50-ns molecular dynamics simulations. This study also introduced our new version of PyPLIF HIPPOS, PyPLIF HIPPOS 0.2.0, which allowed us to identify protein-ligand interaction fingerprints and interaction hotspots resulting from molecular dynamics simulations. Our findings revealed that caffeic acid inhibited the DPP-4 and MMP-9 activity with an IC50 of 158.19 ± 11.30 µM and 88.99 ± 3.35 µM while ethanolic extract of spent coffee grounds exhibited an IC50 of 227.87 ± 23.80 µg/100 µL and 81.24 ± 6.46 µg/100 µL, respectively. Molecular dynamics simulations showed that caffeic acid interacted in the plausible allosteric sites of DPP-4 and in the active site of MMP-9. PyPLIF HIPPOS 0.2.0 identified amino acid residues interacting more than 10% throughout the simulation, which were Lys463 and Trp62 in the plausible allosteric site of DPP-4 and His226 in the active site of MMP-9.

Antispasmodic Effect of Alstonia boonei De Wild. and Its Constituents: Ex Vivo and In Silico Approaches

BACKGROUND

Alstonia boonei, belonging to the family Apocynaceae, is one of the best-known medicinal plants in Africa and Asia. Stem back preparations are traditionally used as muscle relaxants. This study investigated the antispasmodic properties of Alstonia boonei Stem back and its constituents.

METHOD

The freeze-dried aqueous Stem back extract of A. boonei, as well as dichloromethane (DCM), ethyl acetate, and aqueous fractions, were evaluated for their antispasmodic effect via the ex vivo method. Two compounds were isolated from the DCM fraction using chromatographic techniques, and their antispasmodic activity was evaluated. An in silico study was conducted by evaluating the interaction of isolated compounds with human PPARgamma-LBD and human carbonic anhydrase isozyme.

RESULTS

The Stem back crude extract, DCM, ethyl acetate, and aqueous fractions showed antispasmodic activity on high-potassium-induced (K+ 80 mM) contractions on isolated rat ileum with IC50 values of 0.03 ± 0.20, 0.02 ± 0.05, 0.03 ± 0.14, and 0.90 ± 0.06 mg/mL, respectively. The isolated compounds from the DCM fraction were β-amyrin and boonein, with only boonein exhibiting antispasmodic activity on both high-potassium-induced (IC50 = 0.09 ± 0.01 µg/mL) and spontaneous (0.29 ± 0.05 µg/mL) contractions. However, β-amyrin had a stronger interaction with the two proteins during the simulation.

CONCLUSION

The isolated compounds boonein and β-amyrin could serve as starting materials for the development of antispasmodic drugs.

MHC2AffyPred: A machine-learning approach to estimate affinity of MHC class II peptides based on structural interaction fingerprints

Understanding how MHC class II (MHC-II) binding peptides with differing lengths exhibit specific interaction at the core and extended sites within the large MHC-II pocket is a very important aspect of immunological research for designing peptides. Certain efforts were made to generate peptide conformations amenable for MHC-II binding and calculate the binding energy of such complex formation but not directed toward developing a relationship between the peptide conformation in MHC-II structures and the binding affinity (BA) (IC₅₀ ). We present here a machine-learning approach to calculate the BA of the peptides within the MHC-II pocket for HLA-DRA1, HLA-DRB1, HLA-DP, and HLA-DQ allotypes. Instead of generating ensembles of peptide conformations conventionally, the biased mode of conformations was created by considering the peptides in the crystal structures of pMHC-II complexes as the templates, followed by site-directed peptide docking. The structural interaction fingerprints generated from such docked pMHC-II structures along with the Moran autocorrelation descriptors were trained using a random forest regressor specific to each MHC-II peptide lengths (9-19). The entire workflow is automated using Linux shell and Perl scripts to promote the utilization of MHC2AffyPred program to any characterized MHC-II allotypes and is made for free access at https://github.com/SiddhiJani/MHC2AffyPred. The MHC2AffyPred attained better performance (correlation coefficient [CC] of .612-.898) than MHCII3D (.03-.594) and NetMHCIIpan-3.2 (.289-.692) programs in the HLA-DRA1, HLA-DRB1 types. Similarly, the MHC2AffyPred program achieved CC between .91 and .98 for HLA-DP and HLA-DQ peptides (13-mer to 17-mer). Further, a case study on MHC-II binding 15-mer peptides of severe acute respiratory syndrome coronavirus-2 showed very close competency in computing the IC₅₀ values compared to the sequence-based NetMHCIIpan v3.2 and v4.0 programs with a correlation of .998 and .570, respectively.

PyPLIF HIPPOS and Receptor Ensemble Docking Increase the Prediction Accuracy of the Structure-Based Virtual Screening Protocol Targeting Acetylcholinesterase

In this article, the upgrading process of the structure-based virtual screening (SBVS) protocol targeting acetylcholinesterase (AChE) previously published in 2017 is presented. The upgraded version of PyPLIF called PyPLIF HIPPOS and the receptor ensemble docking (RED) method using AutoDock Vina were employed to calculate the ensemble protein–ligand interaction fingerprints (ensPLIF) in a retrospective SBVS campaign targeting AChE. A machine learning technique called recursive partitioning and regression trees (RPART) was then used to optimize the prediction accuracy of the protocol by using the ensPLIF values as the descriptors. The best protocol resulting from this research outperformed the previously published SBVS protocol targeting AChE.

De Novo Molecule Design Using Molecular Generative Models Constrained by Ligand-Protein Interactions

In recent years, molecular deep generative models have attracted much attention for its application in de novo drug design. The data-driven molecular deep generative model approximates the high dimensional distribution of the chemical space through learning from a large number of molecular structural data. So far, most of the molecular generative models rely on purely 2D ligand information in structure generation. Here, we propose a novel molecular deep generative model which adopts a recurrent neural network architecture coupled with a ligand-protein interaction fingerprint as constraints. The fingerprint was constructed on ligand docking poses and represents the 3D binding mode of ligands in the protein pocket. In the current work, generative models constrained with interaction fingerprints were trained and compared with normal RNN models. It has been shown that models trained with constraints of ligand-protein interaction fingerprint have a clear tendency to generating compounds maintaining similar binding modes. Our results demonstrate the potential application of the interaction fingerprint-constrained generative model for the targeted molecule generation and guided exploration on the drug-like chemical space.

ProLIF: a library to encode molecular interactions as fingerprints

Interaction fingerprints are vector representations that summarize the three-dimensional nature of interactions in molecular complexes, typically formed between a protein and a ligand. This kind of encoding has found many applications in drug-discovery projects, from structure-based virtual-screening to machine-learning. Here, we present ProLIF, a Python library designed to generate interaction fingerprints for molecular complexes extracted from molecular dynamics trajectories, experimental structures, and docking simulations. It can handle complexes formed of any combination of ligand, protein, DNA, or RNA molecules. The available interaction types can be fully reparametrized or extended by user-defined ones. Several tutorials that cover typical use-case scenarios are available, and the documentation is accompanied with code snippets showcasing the integration with other data-analysis libraries for a more seamless user-experience. The library can be freely installed from our GitHub repository (https://github.com/chemosim-lab/ProLIF).

PyPLIF HIPPOS-Assisted Prediction of Molecular Determinants of Ligand Binding to Receptors

Identification of molecular determinants of receptor-ligand binding could significantly increase the quality of structure-based virtual screening protocols. In turn, drug design process, especially the fragment-based approaches, could benefit from the knowledge. Retrospective virtual screening campaigns by employing AutoDock Vina followed by protein-ligand interaction fingerprinting (PLIF) identification by using recently published PyPLIF HIPPOS were the main techniques used here. The ligands and decoys datasets from the enhanced version of the database of useful decoys (DUDE) targeting human G protein-coupled receptors (GPCRs) were employed in this research since the mutation data are available and could be used to retrospectively verify the prediction. The results show that the method presented in this article could pinpoint some retrospectively verified molecular determinants. The method is therefore suggested to be employed as a routine in drug design and discovery.