Issue #82: AlphaFold for Docking Screens.

Signal of the Day

AlphaFold for Docking Screens.

AlphaFold is an AI system developed by Google DeepMind to generate three-dimensional structures of proteins without experimental data. The models created with AlphaFold are available on the AlphaFold Protein Structure Database (AlphaFoldDB) ( https://alphafold.ebi.ac.uk/ ). The AlphaFold database is searchable by sequence and protein identification. This chapter focuses on an AlphaFold model and its use for docking screens using Molegro Virtual Docker. We rely on Jupyter Notebooks to integrate docking simulations and build regression models based on the atomic coordinates of protein-pose complexes. Our study focuses on constructing a neural network regression model to predict the inhibition of cyclin-dependent kinase 19 (CDK19). This enzyme is a target for anticancer drugs and does not have experimental data for its atomic coordinates. We utilize the Molegro Data Modeller to construct a regression model based on docking results of inhibitors for which binding affinity data is available. All CDK19 datasets and Jupyter Notebooks discussed in this work are available at GitHub: https://github.com/azevedolab/docking#readme .

Why this matters:

Also Worth Reading

Enhancing CYP450-Ligand Binding Predictions: A Comparative Analysis of Ligand-Based and Hybrid Machine Learning Models.

Predicting cytochrome P450 (CYP450) ligand binding is critical in early-stage drug discovery as CYP450-mediated metabolism profoundly influences drug efficacy, safety, and adverse reaction risks. However, experimental determination of CYP450-ligand interactions remains resource- and time-intensive, underscoring the need for robust computational alternatives. While ligand-based methods are commonly employed, they often fail to fully account for structural intricacies governing protein-ligand interactions. To address this gap, we developed a hybrid machine learning framework integrating ligand descriptors, protein descriptors, and protein-ligand interaction descriptors that include molecular docking-derived parameters, rescoring function components from multiple algorithms, and structural interaction fingerprints (SIFt). Evaluated on CYP1A2 and CYP17A1 isoforms, our model demonstrated superior predictive accuracy in cross-validation compared with stand-alone molecular docking and ligand-based approaches. Furthermore, benchmarking against state-of-the-art tools (SwissADME and ADMETlab 3.0) revealed enhanced performance in binding prediction. This work establishes a versatile framework for advancing computational tools to prioritize CYP450 binding assessments during drug discovery.

Comprehensive Molecular Docking and Molecular Dynamics Reveal Inhibitors of HER2 L755S, T798I, and T798M based on a Large Database of Curcumin Derivatives.

Objective This study presents a methodology employing virtual screening to identify curcumin derivatives with selective affinity for the HER2 mutations L755S, T798I, and T798M. Methods Curcumin derivatives were retrieved from the ChEMBL database and filtered using KNIME. HER2 mutations were modeled in silico using MOE software with PDB ID 3RCD. Molecular docking and dynamics simulations were conducted to screen high-affinity compounds and evaluate binding interactions. Result From 505 curcumin derivatives, the RDKit module implemented in KNIME successfully filtered 317 compounds. Subsequent molecular docking against wild-type HER2 identified 100 curcumin derivatives with low docking scores, among which the top 20 compounds exhibited better binding affinities than Lapatinib. Further molecular docking screening against the three HER2 mutations identified five lead compounds with the lowest docking scores. Molecular docking and molecular dynamics simulation revealed critical binding interactions with residues essential for kinase domain stability. Chemical structural analysis revealed key modifications, such as geranyl and tripeptide modifications. CHEMBL3758656 and CHEMBL3827366, two curcumin derivatives, demonstrated consistent binding across HER2 mutations and a favorable ADMET profile. Conclusion This study successfully identified CHEMBL3758656 and CHEMBL3827366 as promising HER2 inhibitors through comprehensive virtual screening. Their high binding affinity against L755S, T798I, and T798M mutations and favorable ADME and toxicity properties underscore their potential as alternative therapeutics for HER2-positive breast cancer.

Integrative gene target mapping, RNA sequencing, in silico molecular docking, ADMET profiling and molecular dynamics simulation study of marine derived molecules for type 1 diabetes mellitus.

Type 1 diabetes mellitus (T1DM) is a metabolic disease leading threat to human health around the world. Here we aimed to explore new biomarkers and potential therapeutic targets in T1DM through adopting integrated bioinformatics tools. The gene expression Omnibus (GEO) database was used to obtain next generation sequencing data (GSE270484) of T1DM and normal control samples. Furthermore, differentially expressed genes (DEGs) were screened using the DESeq2 package in R bioconductor package. Gene Ontology (GO) and pathway enrichment analyses were performed by g:Profiler. The protein-protein interaction (PPI) network was plotted with IID PPI database and visualized using Cytoscape. Module analysis of the PPI network was done using PEWCC. Then, microRNAs (miRNAs) and transcription factors (TFs) in T1DM were screened out from the miRNet and NetworkAnalyst database. Then, the miRNA-hub gene regulatory network and TF-hub gene regulatory network were constructed by Cytoscape software. Moreover, a drug-hub gene interaction network of the hub genes was constructed and predicted the drug molecule against hub genes. The receiver operating characteristic (ROC) curves were generated to predict diagnostic value of hub genes. Finally we performed molecular docking, ADMET profiling and molecular dynamics simulation studies of marine derived chemical constituents using Schrodinger Suite 2025-1. A total of 958 DEGs were screened: 479 up regulated genes and 479 down regulated genes. DEG were mainly enriched in the terms of developmental process, membrane, cation binding, response to stimulus, cell periphery, ion binding, neuronal system and metabolism. Based on the data of protein-protein interaction (PPI), the top 10 hub genes (5 up regulated and 5 down regulated) were ranked, including FN1, GSN, ADRB2, CEP128, FLNA, CD74, EFEMP2, POU6F2, P4HA2 and BCL6. The miRNA-hub gene regulatory network and TF-hub gene regulatory network showed that hsa-mir-657, hsa-miR-1266-5p, NOTCH1 and GTF3C2 might play an important role in the pathogenesis of T1DM. The drug-hub gene interaction network showed that Clenbuterol, Diethylstilbestrol, Selegiline and Isoflurophate predicted therapeutic drugs for the T1DM. Molecular docking and molecular dynamics simulation study revealed that CMNPD5805 and CMNPD30286 as potential inhibitors of FN1 (pdb id: 3M7P) a key biomarker in pathogenesis of T1DM. These findings promote the understanding of the molecular mechanism and clinically related molecular targets for T1DM.

Research & AI Updates

• Introducing ROCKET: An Upgrade to AlphaFold That Learns From Raw Experimental Data - Simons Foundation — Introducing ROCKET: An Upgrade to AlphaFold That Learns From Raw Experimental Data    Simons Foundation.
• AlphaFold as a prior: experimental structure determination conditioned on a pretrained neural network - Nature — AlphaFold as a prior: experimental structure determination conditioned on a pretrained neural network    Nature.
• Thermo Fisher Unveils New Glacios 3 Cryo-TEM - National Today — Thermo Fisher Unveils New Glacios 3 Cryo-TEM    National Today.
• Thermo Fisher’s Glacios 3 Cryo TEM Puts AI At Center Of Growth - Sahm — Thermo Fisher’s Glacios 3 Cryo TEM Puts AI At Center Of Growth    Sahm.
• Common Ancestry Limits Protein Sequence Exploration, Computational Study Shows - Genetic Engineering and Biotechnology News — Common Ancestry Limits Protein Sequence Exploration, Computational Study Shows    Genetic Engineering and Biotechnology News.

From the Industry

• Tamarind Bio Partners with A-Alpha Bio to Streamline Experimental Validation for AI-Enabled Antibody Discovery - GlobeNewswire — Tamarind Bio Partners with A-Alpha Bio to Streamline Experimental Validation for AI-Enabled Antibody Discovery    GlobeNewswire.
• Unlocking the Potential of Biologics with Advanced Drug Product Manufacturing and Development Technologies - Contract Pharma — Unlocking the Potential of Biologics with Advanced Drug Product Manufacturing and Development Technologies    Contract Pharma.
• Novartis agrees to acquire Excellergy, Inc., building on allergy leadership with next-generation anti-IgE innovation - Novartis — Novartis agrees to acquire Excellergy, Inc., building on allergy leadership with next-generation anti-IgE innovation    Novartis.
• Orionis Biosciences Appoints Dr. Giulio Draetta as Chief Strategy Officer - BioSpace — Orionis Biosciences Appoints Dr.
• Pure Biologics S.A. Stock: Polish Biotech Innovator in Protein Engineering Eyes Global Expansion Ami - AD HOC NEWS — Pure Biologics S.A.
• First batch of foot-ulcer therapy made for planned human trial - Stock Titan — First batch of foot-ulcer therapy made for planned human trial    Stock Titan.

Quick Reads

AlphaFold for Docking Screens.

AlphaFold is an AI system developed by Google DeepMind to generate three-dimensional structures of proteins without experimental data. Read more →

Mapping the inhibition landscape of P-glycoprotein via conformational ensemble docking.

P-glycoprotein (P-gp/ABCB1) is a membrane-bound efflux transporter implicated in multidrug resistance and poor pharmacokinetics of therapeutic agents. Read more →

Synthesis of anticancer agents: molecular docking and ADME studies of new hydropyrimidine linked amide-pyridine derivatives.

To develop novel hydropyrimidine linked amide-pyridine hybrids as anticancer agents. Read more →

Reduced graphene oxide as a potential anticancer nanomaterial: in vitro activity against melanoma and glioblastoma combined with molecular docking insights.

Melanoma and glioblastoma are among the most aggressive cancer types, presenting high recurrence rates, therapeutic resistance, and poor prognosis despite conventional approaches. Read more →

Multi-Target Antidiabetic Potentials of Xylocarpus mekongensis: In Vivo Efficacy, Enzyme Inhibition, and Molecular Docking.

Xylocarpus mekongensis Pierre (Meliaceae), locally known as “Poshur” is a mangrove plant traditionally used in South and Southeast Asia for the management of diabetes and related disorders. Read more →

Tandem Photocatalytic H2O2 Production and In Situ Upgrading Enabled by Docking and Locking Engineered Covalent Organic Frameworks.

Photocatalytic H2O2 generation is hindered by low concentrations and energy-intensive purification, posing major barriers to practical application. Read more →

Transforming a fragile protein helix into an ultrastable scaffold via a hierarchical AI and chemistry framework.

The rational design of proteins that maintain structural integrity under concurrent thermal, mechanical, and chemical stress remains a challenge in molecular engineering. Read more →

SELFormer-guided discovery of xanthohumol and cirsilineol as multi-target natural therapeutics for type 2 diabetes: computational prediction and experimental validation.

Type 2 diabetes mellitus (T2DM) requires multi-target therapeutic approaches addressing both insulin resistance and insulin secretion deficits. Read more →

Pipeline Tip

Employ HADDOCK for ambiguous restraints in protein-protein docking.

Resources & Tools

• Dataset: SCOPe - Curated structural classification of proteins for fold analysis.
• Dataset: Pfam - Protein families database with curated multiple sequence alignments.
• Tool: ChimeraX - Next-gen molecular visualization for large data sets. View all tools →
• Tool: AlphaFold2 - Deep learning system for high-accuracy protein structure prediction. View all tools →
• Event: Structural Biology Events (Open)
• Event: Protein Design Hub (LinkedIn Group) (Ongoing)
• Job: The mid-career reset: how to be strategic about your research direction - Nature at Nature Careers

Deep learning is not a magic wand, but a powerful lens for structural biology. — Recep Adiyaman