#2 · Structural bio & protein design
Web servers deliver structure-based modeling, docking, and mutational prediction
The dominant thread is the packaging of structure-based analyses into web-accessible platforms that take PDB or AlphaFold models as input. BilboMD integrates SAXS/SANS data with AlphaFold-guided preparation, MD conformational sampling, and FoXS/MultiFoXS fitting. FoldDelay quantifies translation-driven delays in native contact formation, SPICE performs comparative analysis of complex interfaces, and i-gRINN supports energy network analysis of GROMACS trajectories with natural language exploration. CB-Dock3 extends blind docking with user-defined regions and metal-aware options, while SGGly and PythiaStudio apply ProtBERT-based and Pythia-based prediction to glycosylation sites and mutational effects. The preprints extend this toward metal-binding site prediction with side-chain descriptors in BioMetAll v2.0 and HDX-MS-guided docking in MInt-HDX. A 14-method benchmark indicates context-dependent performance for remote homology search across sequence, structure, and representation methods.
Published papers
BilboMD: a web-accessible SAXS and AlphaFold-guided modeling pipeline.
BilboMD is a web platform integrating SAXS/SANS data with AlphaFold-guided input prep, MD conformational sampling, FoXS/MultiFoXS fitting, and containerized…
FoldDelay web server: an online tool to quantify translation-driven delays in protein native contact formation.
FoldDelay is a free web server that computes native contact formation delays during translation from AlphaFold or PDB structures, with linked interactive…
PythiaStudio: a one-stop protein engineering platform powered by Pythia model suite.
PythiaStudio is a web platform integrating Pythia, Pythia-PPI, and Pythia-Pocket models to predict mutational effects on stability, binding affinity, and…
Comparative structural analysis of protein complexes with SPICE.
SPICE is a web platform for modular, comparative structural analysis of protein complexes from PDB files, covering interface mapping, bonds, solvent…
CB-Dock3: an enhanced web server for protein-ligand blind docking.
CB-Dock3 is an updated web server for protein-ligand blind docking with a refined engine, expanded templates, user-defined docking regions, and metal-aware…
SGGly: a web server for whole-protein, structure-guided analysis of candidate N-linked glycosylation sites.
SGGly is a web server for whole-protein, structure-guided prediction of N-linked glycosylation sites, using ProtBERT embeddings with sequon and structural…
i-gRINN: a next-generation web server for protein energy network analysis of heterogeneous biomolecular systems with natural language-based data exploration.
i-gRINN is a web server for protein energy network analysis of GROMACS trajectories and PDB ensembles, adding ligand support and LLM-based natural language…
Benchmarking protein sequence and structure search methods for remote homology detection.
Standardized benchmark of 14 protein sequence, structure, and representation-based search methods for remote homology, showing context-dependent performance…
Preprints
BioMetAll v2.0: Introducing Scores, Metal Discrimination, and Side-Chain Descriptors for Predicting Metal-Binding Sites in Proteins.
BioMetAll v2.0 adds metal-specific scoring and side-chain (first coordination sphere) descriptors to predict metal-binding sites in proteins, with improved…
MInt-HDX: Leveraging Hydrogen-Deuterium Exchange Mass Spectrometry and Machine-Learning to Improve Protein-Ligand Docking.
MInt-HDX combines XGBoost on HDX-MS peptide signatures with physics-based docking to generate HDX-guided candidate sites and rank ligand poses, reaching…
Topic history
| Date | Rank |
|---|---|
| #2 | |
| #1 | |
| #4 | |
| #12 | |
| #17 | |
| #33 | |
| #22 | |
| #23 | |
| #28 | |
| #17 | |
| #30 | |
| #30 | |
| #32 |