The IMEx Coronavirus interactome: an evolving map of Coronaviridae-Host molecular interactions

Abstract

The current Coronavirus Disease 2019 (COVID-19) pandemic, caused by the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), has spurred a wave of research of nearly unprecedented scale. Among the different strategies that are being used to understand the disease and develop effective treatments, the study of physical molecular interactions enables studying fine-grained resolution of the mechanisms behind the virus biology and the human organism response. Here we present a curated dataset of physical molecular interactions, manually extracted by IMEx Consortium curators focused on proteins from SARS-CoV-2, SARS-CoV-1 and other members of the Coronaviridae family. Currently, the dataset comprises over 2,200 binarized interactions extracted from 86 publications. The dataset can be accessed in the standard formats recommended by the Proteomics Standards Initiative (HUPO-PSI) at the IntAct database website ( www.ebi.ac.uk/intact ), and will be continuously updated as research on COVID-19 progresses.

Figure 1:. Timeline showing data captured in IMEx resources since COVID-19 outbreak (March 2020).

(A) Cumulative interactions, mutation features and binding regions annotated for SARSCoV-2 (red), SARS-CoV1 (orange) and other Coronaviridae family members proteins (grey). Interactions include spoke-expanded binary relationships. Dots represent the date when the interaction was curated. (B) Amount of cumulative experimental evidence associated with unique binary pairs, captured over time, in each of the three dataset: SARS-CoV-1, SARSCoV-2 and other Coronaviridae family members. Interactions include spoke-expanded binary relationships.

Figure 2:. SARS-CoV-2 HT datasets comparison

SARS-CoV-2 datasets include Gordon_LT (Gordon plus low throughput) and Li_LT (Li plus low throughput). (A) Distribution of HT/LT-derived interactions in the Coronavirus dataset. The number on top of each bar indicates the number of publications per category. High-throughput publications are defined as those hosting more than 50 unique interacting pairs. (B-D) Overlap of viral proteins (B), human targets (C) and viral to human edges (D) across datasets. (E) Distribution of number of interactions between viral proteins and human targets. In the y-scale, 1 indicates that one human protein interacts only with one viral protein, while 16 indicates that one human protein interacts with 16 viral proteins.

Figure 3:. Pathway enrichment analysis of SARS-CoV-1, SARS-CoV-2 and Gordon_LT (Gordon plus low throughput) and Li_LT (Li plus low throughput) datasets.

Enrichment was performed using pathDIP (a database integrating 24 different pathway databases). Only human proteins were considered. The majority of enriched pathways were from Reactome database, so a mapping of each Reactome pathway to the parent pathway ontology was performed, and the heatmap shows the percentage of pathways in each parent pathway over the total of pathways.