Posted in Data visualisation, KEA3

KEA3: A web-based tool to predict involvement of upstream kinases based on a list of gene or protein queries.

(Left) Top 10 kinases that could be involved in transcriptional responses to the yellow fever live-attenuated vaccine (YF-17D). (Right) Interactive map showing the relationship of the different kinases identified in the left panel. Query list is based on the upregulated differentially expressed genes at day 7 post YF-17D vaccination from Querec et al., Nature Immunology, 2009.

Protein kinases catalyze the transfer of a phosphate group from ATP to other proteins’ threonine, serine, or tyrosine residues. This addition of phosphate group to a protein can influence substrate protein activity, stability, localization, and interactions with other molecules. While kinases can be suitably targeted by drugs, characterization of the cell kinome is not easy, as intercellular staining with phospho-specific antibodies is required.

A plausible solution is to leverage on the transcriptomic data to down-select kinase candidates for further validation. Here, I introduce Kinase Enrichment Analysis 3 (KEA3) developed by MV Kuleshov et al., which is a webserver application that predicts upstream kinases based on a list of gene or protein queries.

The KEA3 background database contains measured and predicted kinase-substrate interactions (KSI), kinase-protein interactions (KPI), and interactions supported by co-expression and co-occurrence data. By integrating KSIs and KPIs across data sources, KEA3 produces a composite ranking that improves the recovery of the expected kinases. In addition, the relationship between the top predicted kinase are also displayed in an interactive map.

I tested the ability for KEA3 to evaluate the possible kinases involved in the host transcriptomic responses to the YF-17D vaccine published by Querec et al., Nature Immunology, 2009. Taking the up-regulated differentially expressed genes at day 7 post-YF17D administration as the query list, the top 10 kinase hits are displayed in the figure at the top of this post. Notably, these kinases appear to be highly interconnected and the predicted involvement of EIF2AK2-JAK1-JAK2-TYK axis suggests the involvement of these kinases in triggering type-I interferon responses. This finding is consistent with previous studies showing that YF-17D induces strong interferon and antiviral responses.

Overall, KEA3 is a user-friendly tool that allows users to quickly predict the upstream kinases involved, based on a list of proteins or genes. While an experimental validation will be needed to confirm the involvement of these predicted kinases, the tool provides an informed prediction on the kinases involved that can be used for future studies.