ProteinsPlus: a publicly available resource for protein structure mining @NAR_Open
1. ProteinsPlus is an open-access web server offering a comprehensive suite of tools for protein structure analysis, protein-ligand modeling, and PDB mining—making it one of the most versatile resources for structure-based drug design.
2. The latest update introduces DoGSite3, a robust and orientation-invariant binding pocket predictor with ligand-biased mode, outperforming existing geometry-based tools in pocket detection speed and precision.
3. The GeoMine tool was enhanced with a novel 2D query interface, enabling users to define structural search templates from ligand-bound pockets using intuitive chemical diagrams while synchronizing with a 3D viewer for spatial accuracy.
4. PoseEdit, an upgraded 2D interaction diagram editor, allows interactive exploration and customization of protein-ligand binding modes, introducing new features like ionic and covalent bond detection, color gradients, and synchronized 3D mapping.
5. The new LifeSoaks module analyzes solvent channels in crystallographic structures, enabling researchers to assess accessibility of binding pockets in protein crystals, a key consideration in fragment soaking experiments.
6. ProteinsPlus supports PDB files, AlphaFold predictions, or user-uploaded structures as input, and provides programmatic REST API access and KNIME integration, allowing large-scale, automated workflows for high-throughput structural mining.
7. In a real-world application, the platform was used to mine CK1α binding pockets, perform similarity searches across the PDB using SIENA, assess ligand activity with ActivityFinder, and generate interaction diagrams with PoseEdit—all within a KNIME pipeline.
8. The server’s modular tools are now being redesigned using Django and Vue.js to improve scalability, interoperability, and long-term maintainability—enabling seamless tool chaining like GeoMine-to-SIENA query propagation.
9. From metal coordination (METALizer) and structure quality (StructureProfiler, EDIAscorer) to docking automation (JAMDA, Protoss), ProteinsPlus combines analysis, prediction, and mining tools into a single platform for novice and expert users alike.
10. By integrating cutting-edge visualization, interaction modeling, and automation features, ProteinsPlus empowers the structural biology and medicinal chemistry communities to efficiently extract actionable insights from protein structures.
💻Code:
proteins.plus
📜Paper:
academic.oup.com/nar/advance…
#ProteinStructure #DrugDesign #BindingSitePrediction #ProteinsPlus #StructuralBiology #KNIME #Bioinformatics #LigandInteraction #Docking #WebTools #OpenScience