Motivation: Accurate prediction of protein side-chain conformations is necessary to understand protein folding, protein-protein interactions and facilitate de novo protein design.
Results: Here we apply torsional flow matching and equivariant graph attention to develop FlowPacker, a fast and performant model to predict protein side-chain conformations conditioned on the protein sequence and backbone. We show that FlowPacker outperforms previous state-of-the-art baselines across most metrics with improved runtime. We further show that FlowPacker can be used to inpaint missing side-chain coordinates and also for multimeric targets, and exhibits strong performance on a test set of antibody-antigen complexes.
Availability: Code is available at https://gitlab.com/mjslee0921/flowpacker.
Supplementary Information: Supplementary data are available at Bioinformatics online.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1093/bioinformatics/btaf010 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
MVGNN-PPIS: A novel multi-view graph neural network for protein-protein interaction sites prediction based on Alphafold3-predicted structures and transfer learning.
Int J Biol Macromol
January 2025
College of Information Science and Engineering, Northeastern University, China.
Protein-protein interactions (PPI) are crucial for understanding numerous biological processes and pathogenic mechanisms. Identifying interaction sites is essential for biomedical research and targeted drug development. Compared to experimental methods, accurate computational approaches for protein-protein interaction sites (PPIS) prediction can save significant time and costs.
View Article and Find Full Text PDFBackbone Hydration of -Helical Peptides: Hydrogen-Bonding and Surface Hydrophobicity/Hydrophilicity.
Mol Phys
March 2024
Department of Physics, Brooklyn College of the City University of New York, Brooklyn, NY 11210, United States.
The stability of proteins and small peptides depends on the way they interact with the surrounding water molecules. For small peptides, such as -helical polyalanine (polyALA), water molecules can weaken the intramolecular hydrogen-bonds (HB) formed between the peptide backbone O and NH groups which are responsible for the -helix structure. Here, we perform molecular dynamics simulations to study the hydration of polyALA, polyserine (polySER), and other homopolymer peptide -helices at different temperatures and pressures.
View Article and Find Full Text PDFEvaluation of TOCSY mixing for sensitivity-enhancement in solid-state NMR and application of 4D experiments for side-chain assignments of the full-length 30 kDa membrane protein GlpG.
J Biomol NMR
January 2025
Research Unit Molecular Biophysics, Leibniz-Forschungsinstitut für Molekulare Pharmakologie, Robert- Rössle-Straße 10, 13125, Berlin, Germany.
Chemical shift assignments of large membrane proteins by solid-state NMR experiments are challenging. Recent advancements in sensitivity-enhanced pulse sequences, have made it feasible to acquire H-detected 4D spectra of these challenging protein samples within reasonable timeframes. However, obtaining unambiguous assignments remains difficult without access to side-chain chemical shifts.
View Article and Find Full Text PDFMethods for detecting, building, and improving tryptophan mannosylation in glycoprotein structures.
Protein Sci
February 2025
York Structural Biology Laboratory, Department of Chemistry, University of York, York, UK.
Tryptophan mannosylation, the covalent addition of an α-ᴅ-mannose sugar to a tryptophan side chain, is a post-translational modification (PTM) that can affect protein stability, folding, and interactions. Compared to other forms of protein glycosylation, it is relatively uncommon but is affected by conformational anomalies and modeling errors similar to those seen in N- and O-glycans in the Protein Data Bank (PDB). In this work, we report methods for detecting, building, and improving mannose structures linked to tryptophans.
View Article and Find Full Text PDFEmrE is a bacterial membrane-embedded multidrug transporter that functions as an asymmetric homodimer. EmrE is implicated in antibiotic resistance, but is now known to confer either resistance or susceptibility depending on the identity of the small molecule substrate. Here, we report both solution- and solid-state NMR assignments of S64V-EmrE at pH 5.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!