CoChat

Domain-based prediction of the human isoform interactome provides insights into the functional impact of alternative splicing

Open access
Ghadie, Mohamed Ali, Lambourne, Luke, Vidal, Marc, Xia, Yu · 2017 · PLOS Computational Biology

The DOI matches its Crossref record; 61 of the 61 references it cites were found in a registry or an index.

Abstract

Alternative splicing is known to remodel protein-protein interaction networks ("interactomes"), yet large-scale determination of isoform-specific interactions remains challenging. We present a domain-based method to predict the isoform interactome from the reference interactome. First, we construct the domain-resolved reference interactome by mapping known domain-domain interactions onto experimentally-determined interactions between reference proteins. Then, we construct the isoform interactome by predicting that an isoform loses an interaction if it loses the domain mediating the interaction. Our prediction framework is of high-quality when assessed by experimental data. The predicted human isoform interactome reveals extensive network remodeling by alternative splicing. Protein pairs interacting with different isoforms of the same gene tend to be more divergent in biological function, tissue expression, and disease phenotype than protein pairs interacting with the same isoforms. Our prediction method complements experimental efforts, and demonstrates that integrating structural domain information with interactomes provides insights into the functional impact of alternative splicing.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1371/journal.pcbi.1005717

Cites

61 of 61 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig 1 Computational pipeline for the prediction of the human isoform interactome. Domain-domain interactions are mapped onto two human reference interactomes, HI-II-14 [13] and IntAct [11]. The resulting domain-resolved reference interactomes are then used to predict interactions for the alternative isoforms of proteins in these interactomes, resulting in two predicted isoform interactomes.
Fig 1 Computational pipeline for the prediction of the human isoform interactome. Domain-domain interactions are mapped onto two human reference interactomes, HI-II-14 [13] and IntAct [11]. The resulting domain-resolved reference interactomes are then used to predict interactions for the alternative isoforms of proteins in these interactomes, resulting in two predicted isoform interactomes.CC BY · PubMed Central
Fig 2 Gene Ontology (GO) similarity for protein pairs with different types of isoform interaction profiles. (A) Schematic diagram illustrating different types of isoform interaction profiles for a pair of proteins. (B) GO similarity for pairs of proteins interacting with the same subset of isoforms of the same gene, with different subsets of isoforms of the same gene, and with protein products of different genes in the predicted HI-II-14 isoform interactome. (C) GO similarity for pairs of proteins interacting with the same subset of isoforms of the same gene, with different subsets of isoforms of the same gene, and with protein products of different genes in the predicted IntAct isoform interactome. GO similarity was calculated using the Jaccard similarity index. Pairs in which both proteins have no GO annotations were excluded. Error bars represent standard errors of the mean. Statistical significance was calculated using a two-sided bootstrap test (*: p < 10−2, **: p < 10−3, ***: p < 10−5).
Fig 2 Gene Ontology (GO) similarity for protein pairs with different types of isoform interaction profiles. (A) Schematic diagram illustrating different types of isoform interaction profiles for a pair of proteins. (B) GO similarity for pairs of proteins interacting with the same subset of isoforms of the same gene, with different subsets of isoforms of the same gene, and with protein products of different genes in the predicted HI-II-14 isoform interactome. (C) GO similarity for pairs of proteins interacting with the same subset of isoforms of the same gene, with different subsets of isoforms of the same gene, and with protein products of different genes in the predicted IntAct isoform interactome. GO similarity was calculated using the Jaccard similarity index. Pairs in which both proteins have no GO annotations were excluded. Error bars represent standard errors of the mean. Statistical significance was calculated using a two-sided bootstrap test (*: p < 10−2, **: p < 10−3, ***: p < 10−5).CC BY · PubMed Central
Fig 3 Fraction of shared disease subnetworks for protein pairs with different types of isoform interaction profiles. (A) Fraction of shared disease subnetworks for pairs of proteins interacting with the same subset of isoforms of the same gene, with different subsets of isoforms of the same gene, and with protein products of different genes in the predicted HI-II-14 isoform interactome. (B) Fraction of shared disease subnetworks for pairs of proteins interacting with the same subset of isoforms of the same gene, with different subsets of isoforms of the same gene, and with protein products of different genes in the predicted IntAct isoform interactome. Fraction of shared disease subnetworks was calculated using the Jaccard similarity index. Only pairs in which both proteins are associated with at least one disease subnetwork were considered. A protein belongs to a disease subnetwork if that protein or any of its first-degree neighbors in the HI-II-14 reference binary interactome are associated with the disease. Error bars represent standard errors of the mean. Statistical significance was calculated using a two-sided bootstrap test.
Fig 3 Fraction of shared disease subnetworks for protein pairs with different types of isoform interaction profiles. (A) Fraction of shared disease subnetworks for pairs of proteins interacting with the same subset of isoforms of the same gene, with different subsets of isoforms of the same gene, and with protein products of different genes in the predicted HI-II-14 isoform interactome. (B) Fraction of shared disease subnetworks for pairs of proteins interacting with the same subset of isoforms of the same gene, with different subsets of isoforms of the same gene, and with protein products of different genes in the predicted IntAct isoform interactome. Fraction of shared disease subnetworks was calculated using the Jaccard similarity index. Only pairs in which both proteins are associated with at least one disease subnetwork were considered. A protein belongs to a disease subnetwork if that protein or any of its first-degree neighbors in the HI-II-14 reference binary interactome are associated with the disease. Error bars represent standard errors of the mean. Statistical significance was calculated using a two-sided bootstrap test.CC BY · PubMed Central
Fig 4 Tissue co-expression for protein pairs with different types of isoform interaction profiles. (A) Tissue co-expression for pairs of proteins interacting with the same subset of isoforms of the same gene, with different subsets of isoforms of the same gene, and with protein products of different genes in the predicted HI-II-14 isoform interactome. (B) Tissue co-expression for pairs of proteins interacting with the same subset of isoforms of the same gene, with different subsets of isoforms of the same gene, and with protein products of different genes in the predicted IntAct isoform interactome. Tissue co-expression was calculated using Pearson’s correlation coefficient. Only pairs in which both proteins with expression levels simultaneously quantified in at least 8 tissues were considered. Error bars represent standard errors of the mean. Statistical significance was calculated using a two-sided bootstrap test.
Fig 4 Tissue co-expression for protein pairs with different types of isoform interaction profiles. (A) Tissue co-expression for pairs of proteins interacting with the same subset of isoforms of the same gene, with different subsets of isoforms of the same gene, and with protein products of different genes in the predicted HI-II-14 isoform interactome. (B) Tissue co-expression for pairs of proteins interacting with the same subset of isoforms of the same gene, with different subsets of isoforms of the same gene, and with protein products of different genes in the predicted IntAct isoform interactome. Tissue co-expression was calculated using Pearson’s correlation coefficient. Only pairs in which both proteins with expression levels simultaneously quantified in at least 8 tissues were considered. Error bars represent standard errors of the mean. Statistical significance was calculated using a two-sided bootstrap test.CC BY · PubMed Central
Fig 5 Case studies of protein pairs interacting with different subsets of isoforms of the same gene. (A) Two proteins (EGF and GRB2) interacting with different subsets of isoforms of EGFR. EGF interacts with all four isoforms of EGFR on their ligand-binding domain, whereas GRB2 interacts only with the first isoform (P00533) and does not interact with any of the other three isoforms (P00533-2, P00533-3 and P00533-4) that lack the protein kinase domain. (B) Two proteins (XIAP and APAF1) interacting with different subsets of isoforms of CASP9. APAF1 interacts with three isoforms of CASP9 (P55211, P55211-2 and P55211-3) on their CARD domain and loses interaction with the fourth isoform P55211-4. XIAP interacts with three isoforms of CASP9 (P55211, P55211-2 and P55211-4) on their catalytic domain and has no experimentally confirmed interaction with the third isoform P55211-3.
Fig 5 Case studies of protein pairs interacting with different subsets of isoforms of the same gene. (A) Two proteins (EGF and GRB2) interacting with different subsets of isoforms of EGFR. EGF interacts with all four isoforms of EGFR on their ligand-binding domain, whereas GRB2 interacts only with the first isoform (P00533) and does not interact with any of the other three isoforms (P00533-2, P00533-3 and P00533-4) that lack the protein kinase domain. (B) Two proteins (XIAP and APAF1) interacting with different subsets of isoforms of CASP9. APAF1 interacts with three isoforms of CASP9 (P55211, P55211-2 and P55211-3) on their CARD domain and loses interaction with the fourth isoform P55211-4. XIAP interacts with three isoforms of CASP9 (P55211, P55211-2 and P55211-4) on their catalytic domain and has no experimentally confirmed interaction with the third isoform P55211-3.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Cited by

See all 39

Metadata

Authors
Ghadie, M. A., Lambourne, L., Vidal, M., & Xia, Y.
Affiliations
  • McGill University (CA)
  • Harvard University (US)
  • Dana-Farber Cancer Institute (US)
Corresponding author
Xia, Yu
Venue
PLOS Computational Biology
Published
2017-08-28
Volume
13
Issue
8
Pages
e1005717
ISSN
  • 1553-734X (ISSN-L)
  • 1553-7358
OpenAlex
W2749136581
Cited by
39
References
61
Citations by year
2026: 4 · 2025: 2 · 2024: 6 · 2023: 2 · 2022: 6 · 2021: 5 · 2020: 8 · 2019: 5 · 2018: 1
FWCI
1.68
Type
journal article
Language
en
Open access
Locations
  • PLoS Computational Biology (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Digital Access to Scholarship at Harvard (DASH) (Harvard University) (submittedVersion, cc-by)
  • RePEc: Research Papers in Economics (submittedVersion)
  • eScholarship@McGill (McGill) (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • eScholarship@McGill (McGill) (submittedVersion, cc-by)
  • Figshare (submittedVersion, cc-by)
Funders
  • Natural Sciences and Engineering Research Council of Canada (CA) (RGPIN-2014-03892)
  • Canada Foundation for Innovation (CA) (JELF-33732)
  • Canada Research Chairs (CA)
  • National Human Genome Research Institute (US) (P50HG004233, U01HG001715)
  • National Cancer Institute (R33CA132073)
  • The Ellison Foundation
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Bioinformatics and Genomic Networks
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › RNA Research and Splicing
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Cell Biology › Biotin and Related Studies
Keywords
alternative splicing, protein-protein interaction, protein domains, domain-domain interactions, human interactome, tissue expression, disease phenotypes
MeSH terms
Humans, Neoplasms, Models, Statistical, Alternative Splicing, Computational Biology, Protein Isoforms, Protein Interaction Mapping, Protein Interaction Maps

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.