Integrating protein-protein interactions and text mining for protein function prediction

Abstract

Background: Functional annotation of proteins remains a challenging task. Currently the scientific literature serves as the main source for yet uncurated functional annotations, but curation work is slow and expensive. Automatic techniques that support this work are still lacking reliability. We developed a method to identify conserved protein interaction graphs and to predict missing protein functions from orthologs in these graphs. To enhance the precision of the results, we furthermore implemented a procedure that validates all predictions based on findings reported in the literature.

Results: Using this procedure, more than 80% of the GO annotations for proteins with highly conserved orthologs that are available in UniProtKb/Swiss-Prot could be verified automatically. For a subset of proteins we predicted new GO annotations that were not available in UniProtKb/Swiss-Prot. All predictions were correct (100% precision) according to the verifications from a trained curator.

Conclusion: Our method of integrating CCSs and literature mining is thus a highly reliable approach to predict GO annotations for weakly characterized proteins with orthologs.

Figure 1

Example CCS detected between H. sapiens and S. cerevisiae. The figure shows a conserved and connected subgraph between H. sapiens (circle) and S. cerevisiae (hexagon). Proteins of both species are involved in mRNA splicing and are known to exhibit splicing factor activity to bind the mRNA and support the splicing process. (Solid lines represent conserved PPIs within a species and dashed lines indicate orthology relationships between proteins.)

Figure 2

Schematic illustration for comparing GO annotations in this study. The flowchart summarizes all four combinations for comparing GO annotations of different resources. Protein – annotation associations were extracted from text with or without species identification and GO annotations from text and the UniProtKb/Swiss-Prot database were compared based on exact versus relative matching.

Figure 3

Schematic overview of the studied protein annotations. Schematic overview of the different sets of protein – annotation associations considered in this study.