Proteome-wide assessment of human interactome as a source of capturing domain-motif and domain-domain interactions

Abstract

Protein-protein interactions (PPIs) play a crucial role in various biological processes by establishing domain-motif (DMI) and domain-domain interactions (DDIs). While the existence of real DMIs/DDIs is generally assumed, it is rarely tested; therefore, this study extensively compared high-throughput methods and public PPI repositories as sources for DMI and DDI prediction based on the assumption that the human interactome provides sufficient data for the reliable identification of DMIs and DDIs. Different datasets from leading high-throughput methods (Yeast two-hybrid [Y2H], Affinity Purification coupled Mass Spectrometry [AP-MS], and Co-fractionation-coupled Mass Spectrometry) were assessed for their ability to capture DMIs and DDIs using known DMI/DDI information. High-throughput methods were not notably worse than PPI databases and, in some cases, appeared better. In conclusion, all PPI datasets demonstrated significant enrichment in DMIs and DDIs (p-value <0.001), establishing Y2H and AP-MS as reliable methods for predicting these interactions. This study provides valuable insights for biologists in selecting appropriate methods for predicting DMIs, ultimately aiding in SLiM discovery.

Keywords: DDIs; DMIs; PPIs; SLiMs; high‐throughput methods; proteome.

FIGURE 1

Identification of known DMIs and DDIs from different datasets. (A) Normalized number of DMIs captured over 1000 randomizations. The Y‐axis represents the normalized number of DMIs, with each bar depicting the real DMIs captured over 1000 randomizations, calculated by subtracting random DMIs from observed DMIs. (B) Overlap of captured known DMIs between high‐throughput datasets. (C) Overlap of captured known DMIs between PPI databases. (D) Normalized number of DDIs captured over 1000 randomizations. The Y‐axis signifies the normalized number of DDIs, and each bar indicates the real DDIs captured over 1000 randomizations, obtained by subtracting random DDIs from observed DDIs. (E) Overlap of captured known DDIs between high‐throughput datasets. (F) Overlap of captured known DDIs between PPI databases. (G) Total proportion of DMIs and DDIs captured from the known human DMIs (1236 DMIs) and known human DDIs dataset (5589 DDIs). (H) Percentage of PPIs that are known DMIs, DDIs, or both (nonredundant and reviewed proteins only). The Y‐axis depicts the percentage of PPIs that can be explained as DMIs or DDIs. DDI, domain–domain interaction; DMI, Domain Motif Interactions; PPI, Protein–protein interaction.

FIGURE 2

Predicting DMIs utilizing ELMc‐Protein and ELMc‐Domain strategies. (A) Normalized number of DMIs captured over 1000 randomizations using the ELMc‐Protein strategy. (B) Total proportion of DMIs predicted from potential DMIs using the ELMc‐Protein strategy. Potential DMIs represent the overall proportion of those DMIs that could theoretically be identified based on the proteins in the PPIs. (C) Normalized number of DMIs captured over 1000 randomizations using the ELMc‐Domain strategy. (D) Total proportion of predicted DMIs captured from potential DMIs using the ELMc‐Domain strategy. DMI, Domain Motif Interactions; ELM, Eukaryotic Linear Motif; ELMc‐Domain, Protein Interactions via ELM Class and Pfam Domains; ELMc‐Protein, Known ELM Instances Interacting Proteins.

FIGURE 3

Impact of PPI quality and ELM types. (A) Impact of PPI quality on DMI and DDI enrichment: The X‐axis depicts the confidence scores of various PPI subsets sourced from the HIPPIE database (ranging from 0.1 to 1), while the Y‐axis represents the enrichment score of PPIs within distinct subsets based on their confidence scores. (B) Influence of ELM types on DMI enrichment in interactions from three high‐throughput methods (AP‐MS, Y2H, and CoFrac‐MS) obtained from PPI databases. (C) The normalized count of real DMIs captured by different ELM classes. CoFrac‐MS, Co‐Fraction‐coupled Mass Spectrometry; DDI, domain–domain interaction; DMI, Domain Motif Interactions; ELM, Eukaryotic Linear Motif; HIPPIE, Human Integrated Protein–Protein Interaction rEference; PPI, Protein–protein interaction; Y2H, Yeast two‐hybrid.