ProtCID: A data resource for structural information on protein interactions

Xu, Qifang; Dunbrack, Roland L.

doi:10.1101/579862

Cited by 10 publications

(20 citation statements)

References 101 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, we also feel it is important to provide the biological assembly files. More than half of crystal structures in the PDB have annotated assemblies that are different from the asymmetric unit [ 20 ]. While RCSB does not provide these files in mmCIF format at this time, they are available from PDBe.…”

Section: Resultsmentioning

confidence: 99%

PDBrenum: A webserver and program providing Protein Data Bank files renumbered according to their UniProt sequences

Faezov

Dunbrack

2021

PLoS ONE

Self Cite

View full text Add to dashboard Cite

The Protein Data Bank (PDB) was established at Brookhaven National Laboratories in 1971 as an archive for biological macromolecular crystal structures. In mid 2021, the database has almost 180,000 structures solved by X-ray crystallography, nuclear magnetic resonance, cryo-electron microscopy, and other methods. Many proteins have been studied under different conditions, including binding partners such as ligands, nucleic acids, or other proteins; mutations, and post-translational modifications, thus enabling extensive comparative structure-function studies. However, these studies are made more difficult because authors are allowed by the PDB to number the amino acids in each protein sequence in any manner they wish. This results in the same protein being numbered differently in the available PDB entries. For instance, some authors may include N-terminal signal peptides or the N-terminal methionine in the sequence numbering and others may not. In addition to the coordinates, there are many fields that contain structural and functional information regarding specific residues numbered according to the author. Here we provide a webserver and Python3 application that fixes the PDB sequence numbering problem by replacing the author numbering with numbering derived from the corresponding UniProt sequences. We obtain this correspondence from the SIFTS database from PDBe. The server and program can take a list of PDB entries or a list of UniProt identifiers (e.g., “P04637” or “P53_HUMAN”) and provide renumbered files in mmCIF format and the legacy PDB format for both asymmetric unit files and biological assembly files provided by PDBe.

show abstract

Section: Resultsmentioning

confidence: 99%

PDBrenum: A webserver and program providing Protein Data Bank files renumbered according to their UniProt sequences

Faezov

Dunbrack

2021

PLoS ONE

Self Cite

View full text Add to dashboard Cite

show abstract

“…Firstly, trRosetta models for ∼7,000 Pfam classifications were recently produced and used without modification. Secondly, of the remaining 60% of entries, we used the ProtCID database 30 (http://dunbrack2.fccc.edu/ProtCiD/default.aspx) to link Pfam IDs to known protein structures in the PDB. These exemplar structures from the PDB were downloaded and single chains were extracted from each model (that is, only a single copy of each domain was considered).…”

Section: Methodsmentioning

confidence: 99%

Mining folded proteomes in the era of accurate structure prediction

Bayly-Jones

Whisstock

2021

Preprint

View full text Add to dashboard Cite

Protein structure fundamentally underpins the function and processes of numerous biological systems. Fold recognition algorithms offer a sensitive and robust tool to detect structural, and thereby functional, similarities between distantly related homologs. In the era of accurate structure prediction owing to advances in machine learning techniques, previously curated sequence databases have become a rich source of biological information. Here, we use bioinformatic fold recognition algorithms to scan the entire AlphaFold structure database to identify novel protein family members, infer function and group predicted protein structures. As an example of the utility of this approach, we identify novel, previously unknown members of various pore-forming protein families, including MACPFs, GSDMs and aerolysin-like proteins. Further, we explore the use of structure-based mining for functional inference.

show abstract

“…We followed a procedure described recently. 24 Any protein chains with length less than 30 amino acids are defined as peptides. A Pfam-peptide interface is defined with at least 10 Cb-Cb distances < 12 Å, or ≥ 5 atomic contacts with distance < 5 Å.…”

Section: Peptide-binding Domain Identificationmentioning

confidence: 99%

“…It contains over 14,000 unique domain-domain interfaces and has also recently been expanded to include domain-peptide, domain-ligand, and domain-DNA/RNA interfaces. 24 Here, we present a domain-based approach using biological interfaces from the ProtCID database to identify structure-supported PPI within the human interaction network. First, we methodically assign Pfam (protein family) domains to the human proteome using an updated version of the multi-step greedy algorithm we developed for our previous study.…”

Section: Introductionmentioning

confidence: 99%

Protein domain-based structural interfaces help interpret biologically-relevant interactions in the human interaction network

Kilambi

Gururaj

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

A high-quality map of the human protein-protein interaction (PPI) network can help us better understand complex genotype-phenotype relationships. Each edge between two interacting proteins supported through an interface in a three-dimensional (3D) structure of a protein complex adds credibility to the biological relevance of the interaction. Such structure-supported interactions would augment an interaction map primarily built using high-throughput cell-based biophysical methods. Here, we integrate structural information with the human PPI network to build the structure-supported human interactome, a subnetwork of PPI between proteins that contain domains or regions known to form interfaces in the 3D structures of protein complexes. We expand the coverage of our structure-supported human interactome by using Pfam-based domain definitions, whereby we include homologous interactions if a human complex structure is unavailable. The structure-supported interactome predicts one-eighth of the total network PPI to interact through domain-domain interfaces. It identifies with higher resolution the interacting subunits in multi-protein complexes and enables us to characterize functional and diseaserelevant neighborhoods in the network map with higher accuracy, allowing for structural insights into disease-associated genes and pathways. We expand the structural coverage beyond domain-domain interfaces by identifying the most common non-enzymatic peptide-binding domains with structural support. Adding these interactions between protein domains on one side and peptide regions on the other approximately doubles the number of structure-supported PPI. The human structure-supported interactome is a resource to prioritize investigations of smaller-scale context-specific experimental PPI neighborhoods of biological or clinical significance. Short abstractA high-quality map of the human protein-protein interaction (PPI) network can help us better understand genotype-phenotype relationships. Each edge between two interacting proteins supported through an interface in a three-dimensional structure of a protein complex adds credibility to the biological relevance of the interaction aiding experimental prioritization. Here, we integrate structural information with the human interactome to build the structure-supported human interactome, a subnetwork of PPI between proteins that contain domains or regions known to form interfaces in the structures of protein complexes. The structure-supported interactome predicts one-eighth of the total PPI to interact through domaindomain interfaces. It identifies with higher resolution the interacting subunits in multi-protein complexes and enables us to structurally characterize functional, disease-relevant network neighborhoods. We also expand the structural coverage by identifying PPI between non-enzymatic peptide-binding domains on one side and peptide regions on the other, thereby doubling the number of structure-supported PPI.

show abstract

ProtCID: A data resource for structural information on protein interactions

Cited by 10 publications

References 101 publications

PDBrenum: A webserver and program providing Protein Data Bank files renumbered according to their UniProt sequences

PDBrenum: A webserver and program providing Protein Data Bank files renumbered according to their UniProt sequences

Mining folded proteomes in the era of accurate structure prediction

Protein domain-based structural interfaces help interpret biologically-relevant interactions in the human interaction network

Contact Info

Product

Resources

About