3D Structure Prediction Research Articles

Has AlphaFold3 achieved success for RNA?

Predicting the 3D structure of RNA is a significant challenge despite ongoing advancements in the field. Although AlphaFold has successfully addressed this problem for proteins, RNA structure prediction raises difficulties due to the fundamental differences between proteins and RNA, which hinder its direct adaptation. The latest release of AlphaFold, AlphaFold3, has broadened its scope to include multiple different molecules such as DNA, ligands and RNA. While the AlphaFold3 article discussed the results for the last CASP-RNA data set, the scope of its performance and the limitations for RNA are unclear. In this article, we provide a comprehensive analysis of the performance of AlphaFold3 in the prediction of 3D structures of RNA. Through an extensive benchmark over five different test sets, we discuss the performance and limitations of AlphaFold3. We also compare its performance with ten existing state-of-the-art ab initio, template-based and deep-learning approaches. Our results are freely available on the EvryRNA platform at https://evryrna.ibisc.univ-evry.fr/evryrna/alphafold3/.

Metacaspases-Like Proteases of Trichomonas vaginalis: In Silico Identification and Characterization.

Metacaspases (MCA), are cysteine-dependent proteases closely related to caspases. In protozoa, MCA plays an important role in programmed cell death (PCD). In Trichomonas vaginalis, a kind of PCD that resembles apoptosis has been described, but the activators of this mechanism have not been demonstrated. We performed a genome-wide in silico analysis in the T. vaginalis database using consensus MCA domains. A total of 15 protein annotations for MCA-like sequences were retrieved. Only 7/15 (TvMCA1-6 and TvMCA9) of the sequences were annotated as putative MCA and exhibited a similar range of amino acid length in comparison to the consensus sequences used for the query. By in silico analysis, we found that they are thermostable, hydrophilic proteins with molecular weights ranging from 27 to 33 KDa and their theoretical isoelectric points are in a 5.08-8.57 range. The phylogenetic analysis showed the similarity of conserved motifs for the predicted TvMCA proteins. 3D structure prediction by homology modeling demonstrated that TvMCA proteins show a similar conformation to crystallized MCA proteins. Taken together, our results indicate that these trichomonad proteins have conserved sequences like MCA proteins and suggest that they may be responsible for proteolytic activity during a PCD-like mechanism in this parasite.

RNA-TorsionBERT: leveraging language models for RNA 3D torsion angles prediction.

Predicting the 3D structure of RNA is an ongoing challenge that has yet to be completely addressed despite continuous advancements. RNA 3D structures rely on distances between residues and base interactions but also backbone torsional angles. Knowing the torsional angles for each residue could help reconstruct its global folding, which is what we tackle in this work. This paper presents a novel approach for directly predicting RNA torsional angles from raw sequence data. Our method draws inspiration from the successful application of language models in various domains and adapts them to RNA. We have developed a language-based model, RNA-TorsionBERT, incorporating better sequential interactions for predicting RNA torsional and pseudo-torsional angles from the sequence only. Through extensive benchmarking, we demonstrate that our method improves the prediction of torsional angles compared to state-of-the-art methods. In addition, by using our predictive model, we have inferred a torsion angle-dependent scoring function, called TB-MCQ, that replaces the true reference angles by our model prediction. We show that it accurately evaluates the quality of near-native predicted structures, in terms of RNA backbone torsion angle values. Our work demonstrates promising results, suggesting the potential utility of language models in advancing RNA 3D structure prediction. Source code is freely available on the EvryRNA platform: https://evryrna.ibisc.univ-evry.fr/evryrna/RNA-TorsionBERT. Supplementary data are available from the website.

An easy-to-use three-dimensional protein-structure-prediction online platform "DPL3D" based on deep learning algorithms.

The change in the three-dimensional (3D) structure of a protein can affect its own function or interaction with other protein(s), which may lead to disease(s). Gene mutations, especially missense mutations, are the main cause of changes in protein structure. Due to the lack of protein crystal structure data, about three-quarters of human mutant proteins cannot be predicted or accurately predicted, and the pathogenicity of missense mutations can only be indirectly evaluated by evolutionary conservation. Recently, many computational methods have been developed to predict protein 3D structures with accuracy comparable to experiments. This progress enables the information of structural biology to be further utilized by clinicians. Thus, we developed a user-friendly platform named DPL3D (http://nsbio.tech:3000) which can predict and visualize the 3D structure of mutant proteins. The crystal structure and other information of proteins were downloaded together with the software including AlphaFold 2, RoseTTAFold, RoseTTAFold All-Atom, and trRosettaX-Single. We implemented a query module for 210,180 molecular structures, including 52,248 human proteins. Visualization of protein two-dimensional (2D) and 3D structure prediction can be generated via LiteMol automatically or manually and interactively. This platform will allow users to easily and quickly retrieve large-scale structural information for biological discovery.

In Silico Optimization of a Streptomyces sp. Antimicrobial Peptide Against Oncobacteria in Oral Squamous Cell Carcinoma

ABSTRACTOral squamous cell carcinoma (OSCC) is a major public health concern, accounting for more than 90% of all oral malignancies. Pathogenic bacteria, such as Streptococcus mutans, Fusobacterium nucleatum, and Porphyromonas gingivalis, are closely linked to OSCC and promote chronic inflammation, a key factor in the tumor microenvironment. The increasing incidence of oral malignancies necessitates the development of antimicrobial peptides (AMPs) with enhanced anticancer activity as targeted therapeutics against bacteria to mitigate their impact on OSCC. A total of 182 Streptomyces genus‐derived AMPs were collected and analyzed in silico to determine their potent antibacterial and anticancer characteristics. Fourteen AMPs were tested for 3D structural alignment prediction and validation, and the most efficient peptide was determined to be sAMP2. A genome‐based analysis of the peptide sourced from Streptomyces globisporus C‐1027 was conducted to explore its evolutionary significance. The cell‐penetrating peptide activity of sAMP2 was optimized to make it more stable and better able to target OSCC host cellular receptors. The peptide Mut_sAMP2 was found to interact with the bacterial virulence proteins DEX of S. mutans, LPS of F. nucleatum, and KGP of P. gingivalis and with OSCC‐associated proteins (TNFα, IL‐6, IL‐8, p38, p53, E‐cadherin, Jak‐1, and PAR2) with greater binding affinity. The peptide Mut_sAMP2 treated OECM‐1 cells showed reduced cell survival by the cytotoxicity activity of IC50 5.4 μg/mL. To the best of our knowledge, our computational study is the first to significantly impact the development of new and effective cancer therapeutics from Streptomyces sp. derived AMPs with a broad spectrum of activity and potential evaluating to combat OSCC progression linked with oral oncobacteria.

Deciphering proteins in Alzheimer's disease: A new Mendelian randomization method integrated with AlphaFold3 for 3D structure prediction.

Hidden confounding biases hinder identifying causal protein biomarkers for Alzheimer's disease in non-randomized studies. While Mendelian randomization (MR) can mitigate these biases using protein quantitative trait loci (pQTLs) as instrumental variables, some pQTLs violate core assumptions, leading to biased conclusions. To address this, we propose MR-SPI, a novel MR method that selects valid pQTL instruments using Leo Tolstoy's Anna Karenina principle and performs robust post-selection inference. Integrating MR-SPI with AlphaFold3, we developed a computational pipeline to identify causal protein biomarkers and predict 3D structural changes. Applied to genome-wide proteomics data from 54,306UK Biobank participants and 455,258 subjects (71,880 cases and 383,378 controls) for a genome-wide association study of Alzheimer's disease, we identified seven proteins (TREM2, PILRB, PILRA, EPHA1, CD33, RET, and CD55) with structural alterations due to missense mutations. These findings offer insights into the etiology and potential drug targets for Alzheimer's disease.

Analysis of the Molecular Structure of Lipase-Dependent Chaperone from Ralstonia pickettii Strain BK6

Several biotechnology industries are exploring the characteristics of lipase-dependent chaperones due to their distinctive biochemical traits. This study aimed to employ bioinformatics to analyze the molecular structure of Ralstonia picketii BK6's lipase-dependent chaperon (LipRM). The sequence mapping and amino acid distribution were examined using BioEdit (version 7.0.9.1). SignalP 5.0 and Interpro are employed for signal peptide detection, whereas Swiss-Model and VMD 1.9.2 are used for molecular dynamics modelling. The results showed that the Shine-Dalgarno sequence was discovered in the LipRM promoter, seven nucleotides upstream of the initiation codon (AUG) with the 5'-AGGAGA-3', and has a terminator region that facilitates the formation of a secondary structure. The protein's 3D structure prediction results indicate differences in the alpha helix chains (residues 166-174 and 254-271) between LipRM and the reference lipase. LipRM's molecular structure comprises a detachable signal peptide, and with variations in helix alpha chain conformation and ligand geometry.

Accurate RNA 3D structure prediction using a language model-based deep learning approach

Accurate prediction of RNA three-dimensional (3D) structures remains an unsolved challenge. Determining RNA 3D structures is crucial for understanding their functions and informing RNA-targeting drug development and synthetic biology design. The structural flexibility of RNA, which leads to the scarcity of experimentally determined data, complicates computational prediction efforts. Here we present RhoFold+, an RNA language model-based deep learning method that accurately predicts 3D structures of single-chain RNAs from sequences. By integrating an RNA language model pretrained on ~23.7 million RNA sequences and leveraging techniques to address data scarcity, RhoFold+ offers a fully automated end-to-end pipeline for RNA 3D structure prediction. Retrospective evaluations on RNA-Puzzles and CASP15 natural RNA targets demonstrate the superiority of RhoFold+ over existing methods, including human expert groups. Its efficacy and generalizability are further validated through cross-family and cross-type assessments, as well as time-censored benchmarks. Additionally, RhoFold+ predicts RNA secondary structures and interhelical angles, providing empirically verifiable features that broaden its applicability to RNA structure and function studies.

Unraveling the physiochemical characteristics and molecular insights of Zein protein through structural modeling and conformational dynamics: a synergistic approach between machine learning and molecular dynamics simulations

This research article presents a comprehensive investigation into the three-dimensional structure, physicochemical characteristics and conformational stability of the Zein protein. Machine learning (ML) based homology modeling approach, was employed to predict the 3D structure of Zein protein. Convolutional neural networks (CNNs) were utilized for refining the model, capturing complex spatial features and improving decoy refinement. The predicted 3D structure of Zein protein showed a high-confidence score, i.e. C-score of 0.96. Physiochemical characteristic was also analyzed to investigate its protonation and deprotonation behavior across a range of pH values. A comprehensive analysis of the titration curve and electrostatic charges was performed to uncover valuable molecular insights into the zein protein’s charge distribution, electrostatic interactions and potential conformational changes. Molecular dynamics (MD) simulations were performed to analyze the zein structural behavior under different pH values (2.0, 4.5, 6.8, 10.0 and 12.5), ionic strengths (0 mM, 25 mM, 50 mM, 75 mM, 100 mM) and temperatures (300K, 350K, 375K). Our results demonstrated the influence of these factors on zein protein’s stability and conformational dynamics. At extreme pH values of 2.0 and 12.5, the Zein protein exhibited increased structural deviations and potential unfolding, while intermediate pH values closer to the protein’s isoelectric point (pI) demonstrated more compact and stable conformations. Analysis of root mean square deviation, radius of gyration, solvent accessible surface area and Ramachandran plot provided clear understandings of the protein’s compactness and surface exposure, confirming the impact of pH, ionic strength and temperature on the protein’s conformation.

IsRNAcirc: 3D structure prediction of circular RNAs based on coarse-grained molecular dynamics simulation.

As an emerging class of RNA molecules, circular RNAs play pivotal roles in various biological processes, thereby determining their three-dimensional (3D) structure is crucial for a deep understanding of their biological significances. Similar to linear RNAs, the development of computational methods for circular RNA 3D structure prediction is challenging, especially considering the inherent flexibility and potentially long length of circular RNAs. Here, we introduce an extension of our previous IsRNA2 model, named IsRNAcirc, to enable circular RNA 3D structure predictions through coarse-grained molecular dynamics simulations. The workflow of IsRNAcirc consists of four main steps, including input preparation, end closure, structure prediction, and model refinement. Our results demonstrate that IsRNAcirc can provide reasonable 3D structure predictions for circular RNAs, which significantly reduce the locally irrational elements contained in the initial input. Moreover, for a validation test set comprising 34 circular RNAs, our IsRNAcirc can generate 3D models with better scores than the template-based 3dRNA method. These findings demonstrate that our IsRNAcirc method is a promising tool to explore the structural details along with intricate interactions of circular RNAs.

Computational-guided discovery of UDP-glycosyltransferases for lauryl glucoside production using engineered E. coli

Defining suitable enzymes for reaction steps in novel synthetic pathways is crucial for developing microbial cell factories for non-natural products. Here, we developed a computational workflow to identify C12 alcohol-active UDP-glycosyltransferases. The workflow involved three steps: (1) assembling initial candidates of putative UDP-glycosyltransferases, (2) refining selection by examining conserved regions, and (3) 3D structure prediction and molecular docking. Genomic sequences from Candida, Pichia, Rhizopus, and Thermotoga, known for lauryl glucoside synthesis via whole-cell biocatalysis, were screened. Out of 240 predicted glycosyltransferases, 8 candidates annotated as glycosyltransferases were selected after filtering out those with signal peptides and identifying conserved UDP-glycosyltransferase regions. These proteins underwent 3D structure prediction and molecular docking with 1-dodecanol. RO3G, a candidate from Rhizopus delemar RA 99–880 with a relatively high ChemPLP fitness score, was selected and expressed in Escherichia coli BL21 (DE3). It was further characterized using a feeding experiment with 1-dodecanol. Results confirmed that the RO3G-expressing strain could convert 1-dodecanol to lauryl glucoside, as quantified by HPLC and identified by targeted LC-MS. Monitoring the growth and fermentation profiles of the engineered strain revealed that RO3G expression did not affect cell growth. Interestingly, acetate, a major fermentation product, was reduced in the RO3G-expressing strain compared to the GFP-expressing strain, suggesting a redirection of flux from acetate to other pathways. Overall, this work presents a successful workflow for discovering UDP-glycosyltransferase enzymes with confirmed activity toward 1-dodecanol for lauryl glucoside production.Graphical abstract

Design and development of dual targeting CAR protein for the development of CAR T-cell therapy against KRAS mutated pancreatic ductal adenocarcinoma using computational approaches

Mutant KRAS promotes the proliferation, metastasis, and aggressiveness of various cancers including pancreatic ductal adenocarcinoma (PDAC), non-small cell lung cancer (NSCLC), and colorectal adenocarcinoma (CRC) respectively. Mutant KRAS therapeutics are limited, while Sotorasib and Adagrasib were the only FDA-approved drugs for the treatment of KRASG12C mutated NSCLC. Chimeric antigen receptor (CAR) T-cell therapy has been emerged as an effective strategy against hematological malignancies and being extended towards solid cancers including PDAC. mesothelin (MSLN) and Carcinoembryonic Antigen (CEA) were reported to be highly overexpressed in KRAS-mutated PDAC. Meanwhile, in clinical trials, several CAR T-cell therapy studies are mainly focused towards these two cancer antigens in PDAC, however, the dual targeting of these two neoantigens is not reported. In the present study, we have designed and developed a novel dual-targeting CAR protein by employing various bioinformatics approaches such as functional analysis (antigenicity, allergenicity, antigen binding sites & signalling cascades), qualitative analysis (physicochemical, prediction, refinement & validation of 2D and 3D structures), molecular docking, and in silico cloning. Our results revealed that the designed CAR protein specifically binds with both MSLN & CEA with significant binding affinities, and was predicted to be stable & non-allergenic. Additionally, the protein–protein interaction network reveals the T-cell mediated antitumor responses of each domain in the designed CAR. Conclusively, we have designed and developed a dual targeting (MSLN & CEA) CAR protein towards KRAS-mutated PDAC using computational approaches. Alongside, we further recommend to engineer this designed CAR in T-cells and evaluating their therapeutic efficiency in in vitro and in vivo studies in the near future.

Dinucleotide composition representation -based deep learning to predict scoliosis-associated Fibrillin-1 genotypes.

Scoliosis is a pathological spine structure deformation, predominantly classified as "idiopathic" due to its unknown etiology. However, it has been suggested that scoliosis may be linked to polygenic backgrounds. It is crucial to identify potential Adolescent Idiopathic Scoliosis (AIS)-related genetic backgrounds before scoliosis onset. The present study was designed to intelligently parse, decompose and predict AIS-related variants in ClinVar database. Possible AIS-related variant records downloaded from ClinVar were parsed for various labels, decomposed for Dinucleotide Compositional Representation (DCR) and other traits, screened for high-risk genes with statistical analysis, and then learned intelligently with deep learning to predict high-risk AIS genotypes. Results demonstrated that the present framework is composed of all technical sections of data parsing, scoliosis genotyping, genome encoding, machine learning (ML)/deep learning (DL) and scoliosis genotype predicting. 58,000 scoliosis-related records were automatically parsed and statistically analyzed for high-risk genes and genotypes, such as FBN1, LAMA2 and SPG11. All variant genes were decomposed for DCR and other traits. Unsupervised ML indicated marked inter-group separation and intra-group clustering of the DCR of FBN1, LAMA2 or SPG11 for the five types of variants (Pathogenic, Pathogeniclikely, Benign, Benignlikely and Uncertain). A FBN1 DCR-based Convolutional Neural Network (CNN) was trained for Pathogenic and Benign/ Benignlikely variants performed accurately on validation data and predicted 179 high-risk scoliosis variants. The trained predictor was interpretable for the similar distribution of variant types and variant locations within 2D structure units in the predicted 3D structure of FBN1. In summary, scoliosis risk is predictable by deep learning based on genomic decomposed features of DCR. DCR-based classifier has predicted more scoliosis risk FBN1 variants in ClinVar database. DCR-based models would be promising for genotype-to-phenotype prediction for more disease types.

The Structure Analysis and mRNA Expression of CaV2 Gene Responding to Hypoxia Stress in Anadara granosa

The blood clam (Anadara granosa) is an economic bivalve that is relatively tolerant to hypoxia, but its molecular mechanism of hypoxia tolerance is unclear. We found that a significant decrease in extracellular Ca2+ concentration and a marked increase in intracellular Ca2+ concentration was observed in the blood clam through the fluorescence probe method, under hypoxic conditions at 0.5 mg/L. Concomitantly, there was a downward trend in the expression level of CaV2 mRNA, whereas NFAT (nuclear factor of activated T cells) expression increased by qRT-PCR. These findings suggest that the elevated intracellular Ca2+ concentration may activate negative transcription factors of NFAT, which subsequently suppresses the transcription of CaV2, leading to its decreased expression. Then, the NFAT RNA interference experiments supported this hypothesis. Sequence analysis and 3D structure prediction revealed conserved and mutated residue sites in blood clam compared to other bivalves. Hypoxia-induced changes in intracellular and extracellular Ca2+ concentrations, activating transcription factor NFAT and suppressing CaV2 expression. This study highlights the key roles of CaV2 and NFAT in hypoxia adaptation, paving the way for further exploration of hypoxia tolerance mechanisms in mollusca.

Experimental Spectroscopic and Computational Studies on A New Synthesized Sulfisoxazole Derivative; Molecular Docking, Drug-likeness, ADME, Toxicity Predictions and Carbonic Anhydrase II Activity Investigations

In this paper, 5-Dimethylamino-2-hydroxy-3-methoxy benzaldehyde sulfisoxazole (5DHMS) and Cu(5DHMS)2 compounds have been synthesized. The molecular structure characterizations were performed using experimental and computational methods. In the experimental part of the study, CHNS, FT-IR, NMR, TGA, and LC-MS analysis techniques were used. In the theoretical part, Density Functional Theory (DFT) was selected for the calculation level. Firstly, optimized structures were obtained from the predicted 3D structures. Vibrational modes were calculated using the optimized structures of the compounds. The vibrational modes of each compound were analyzed in detail using potential energy distribution (PED). Molecular electrostatic potential and HOMO-LUMO maps were drawn. Global chemical reactivity descriptors of compounds were determined. Moreover, the solvent media effects on chemical reactivity descriptors were revealed in detail. Protein-ligand interactions with the target receptor carbonic anhydrase II enzyme were performed. In addition, some important biological activity parameters such as drug-likeness, toxicity, and ADME predictions were examined in silico.

Computational Analysis and Inhibition Study of Carbonic Anhydrase from Azadirachta indica Leaves

Introduction: Carbonic anhydrase (CA) has been an enzyme of great interest for its application in carbon dioxide sequestration. A total of 66 protein sequences of plant CAs were computationally analyzed and submitted to bioinformatics for motif and domain identification, multiple sequence alignment, and phylogenetic analysis. Method: The current work aims to extend knowledge among researchers in better understanding the structure of AZDI CA by analyzing its physicochemical properties, secondary structure prediction, 3D modeling of protein sequence, and its validation using a variety of conventional computational methods. Results: The accuracy of the predicted 3D structure checked by the Ramachandran plot generated by PROCHECK showed that for the CA protein sequence 93.1%, was observed in the most favored regions, VERIFY 3D 69.44% and ERRAT 98.09%. AZDI CA was docked with various anions and sulphonamide derivatives to study their inhibitory effect and calculate binding energy by using Autodock 4.2. Ethoxzolamide is found to be a better inhibitor of AZDI CA with binding energy -8.71 kCal/mol and KI value 0.0004 mM. Further MD simulation of the docked complex of ethoxzolamide with AZDI CA was done using SCHRODINGER DESMOND software. Conclusion: The outcomes of this research will have a significant influence on biochemistry, biotechnology, and potential applications of AZDI CA and similar plant CAs in determining potent inhibitors for drug targets in treating various diseases.

In silico Discovery of Leptukalins, The New Potassium Channel Blockers from the Iranian Scorpion, Hemiscorpius Lepturus.

Blocking Kv 1.2 and Kv 1.3 potassium channels using scorpion venom- derived toxins holds potential therapeutic value. These channels are implicated in autoimmune diseases such as neurodegenerative diseases, multiple sclerosis, rheumatoid arthritis, and type 1 diabetes. The present work aims at the discovery and in silico activity analysis of potassium channel blockers (KTxs) from the cDNA library derived from the venom gland of Iranian scorpion Hemiscorpius lepturus (H. lepturus). The sequence regarding potassium channel blockers were extracted based on Gene Ontology for H. lepturus venom gland. Homology analyses, superfamily, family, and evolutionary signatures of H. lepturus KTxs (H.L KTxs) were determined by using BLASTP, COBALT, PROSITE, and InterPro servers. The predicted 3D structures of H.L KTxs were superimposed against their homologs to predict structure activity relationship. Molecular docking analysis was also performed to predict the binding affinity of H.L KTxs to Kv 1.2 and Kv 1.3 channels. Finally, the toxicity was predicted. Seven H.L KTxs, designated as Leptukalin, were extracted from the cDNA library of H. lepturus venom gland. Homology analyses proved that they can act as potassium channel blockers and they belong to the superfamily and family of Scorpion Toxin-like and Short-chain scorpion toxins, respectively. Structural alignment results confirmed the activity of H.L KTxs. Binding affinity of all H.L KTxs to Kv 1.2 and Kv 1.3 channels ranged from -4.4 to -5.5 and -4 to -5.7 Kcal/mol, respectively. In silico toxicity assay showed that Leptukalin 3, Leptukalin 5, and Leptukalin 7 were non-toxic. Three non-toxic KTxs, Leptukalin 3, 5, and 7, were successfully discovered from the cDNA library of H. lepturus venom gland. Gathering all data together, the discovered peptides are promising potassium channel blockers. Accordingly, Leptukalin 3, 5, and 7 could be suggested for complementary in vitro studies and mouse model of autoimmune diseases.

Reliability of AlphaFold2 Models in Virtual Drug Screening: A Focus on Selected Class A GPCRs.

Protein three-dimensional (3D) structure prediction is one of the most challenging issues in the field of computational biochemistry, which has overwhelmed scientists for almost half a century. A significant breakthrough in structural biology has been established by developing the artificial intelligence (AI) system AlphaFold2 (AF2). The AF2 system provides a state-of-the-art prediction of protein structures from nearly all known protein sequences with high accuracy. This study examined the reliability of AF2 models compared to the experimental structures in drug discovery, focusing on one of the most common protein drug-targeted classes known as G protein-coupled receptors (GPCRs) class A. A total of 32 representative protein targets were selected, including experimental structures of X-ray crystallographic and Cryo-EM structures and their corresponding AF2 models. The quality of AF2 models was assessed using different structure validation tools, including the pLDDT score, RMSD value, MolProbity score, percentage of Ramachandran favored, QMEAN Z-score, and QMEANDisCo Global. The molecular docking was performed using the Genetic Optimization for Ligand Docking (GOLD) software. The AF2 models' reliability in virtual drug screening was determined by their ability to predict the ligand binding poses closest to the native binding pose by assessing the Root Mean Square Deviation (RMSD) metric and docking scoring function. The quality of the docking and scoring function was evaluated using the enrichment factor (EF). Furthermore, the capability of using AF2 models in molecular docking to identify hits with key protein-ligand interactions was analyzed. The posing power results showed that the AF2 models successfully predicted ligand binding poses (RMSD < 2 Å). However, they exhibited lower screening power, with average EF values of 2.24, 2.42, and 1.82 for X-ray, Cryo-EM, and AF2 structures, respectively. Moreover, our study revealed that molecular docking using AF2 models can identify competitive inhibitors. In conclusion, this study found that AF2 models provided docking results comparable to experimental structures, particularly for certain GPCR targets, and could potentially significantly impact drug discovery.

Analysis of Phylogenetics and Structural Relationships of the Cytochrome b Gene Protein in Mahseer Fish (Tor sp.)

Abstract The Cytochrome b gene from mitochondrial DNA encodes a protein that can be used as a genetic marker to identify species. This study aims to determine the bioinformatic characteristics of the gene encoding Cyt b isolated from various Mahseer species, Tor tambroides, Tor douronensis, Tor khudree, and Tor putitora. Mahseer Cyt b sequences can be used for detailed computational investigation of the properties of 3D protein models and phylogenetics. The data analysis technique uses nucleotides with a sequence length of 330 bp obtained from NCBI. The MEGA X program was used for phylogenetic analysis. 3D protein structure predictions were obtained through the Swiss model and protein homology program using PYMOL software. The results of phylogenetics show close kinship relationships with the closest genetic distance of 0.000, while the farthest distance is 0.068. The nucleotide composition contains Thymine (30.23%), Cytosine (25.81%), Arginine (28.05%), and Guanine (15.92%). 3D modeling of the Cyt b protein Mahseer referring to the species Neolissochilus hexagonolepis (McClelland, 1839). The 3D protein structure prediction was assessed via a Ramachandran plot of 99.07%. This study can be used as a reference to support sustainable conservation efforts by observing the populations and habitat conditions.

Development of Drug Discovery Platforms Using Artificial Intelligence and Cheminformatics.

Recently, remarkable progress has been achieved in artificial intelligence (AI), including machine learning. Various AI models have been proposed for drug discovery, including the design of small molecules, activity prediction, and three-dimensional (3D) structure prediction of proteins. AI consists of diverse elements, including information retrieval and machine learning, and can be used in a wide range of drug discovery scenarios. In this review, we focused on AI for small-molecule drug discovery with respect to molecular design, activity prediction, and prediction of the binding poses of compounds to target molecules. We also discussed the applications of AI in academic drug discovery.