Hayai-Annotation v3.0: A functional gene prediction tool that integrates orthologs and gene ontology for network analysis

Andrea Ghelfi,Sachiko Isobe

DOI: https://doi.org/10.1101/2024.06.05.597500

2024-06-06

Abstract:Hayai-Annotation v3, an R-package integrated with the R-Shiny browser interface, utilizes two methods for functional annotation: DIAMOND for sequence alignment using UniProtKB Plants as the database, and OrthoLoger, the official OrthoDB tool for ortholog inferences. The GO enrichment accuracy was assessed by a CAFA-evaluator, demonstrating that Hayai-Annotation v3's accuracy was comparable to that of the benchmark, BLAST2GO. We here propose a method to explore genome evolution and adaptation from a different perspective, by creating networks and heatmaps correlating orthologs with gene ontology (molecular function and biological process) from their co-occurrence tables. This approach enhances the ability to infer functions of uncharacterized genes by associating orthologs with gene ontology terms and the ability to visualize the distribution of gene numbers correlated with co-occurrence patterns across different species. To our knowledge, this is the first attempt to correlate orthologs with GO (MF and BP) to construct a gene network, providing a comprehensive, cross-species view of gene distribution and function. Hayai-Annotation v3 not only retains the convenience of previous versions but also enhances ortholog analysis functionality, allowing for evolutionary insights from gene sequences. Hayai-Annotation v3 is expected to contribute significantly to the future development of plant genome analysis.

Bioinformatics

What problem does this paper attempt to address?

The paper mainly focuses on the problem of predicting plant gene function. The authors developed a R package tool called Hayai-Annotation v3, which combines orthologs and Gene Ontology (GO) information for network analysis. This tool utilizes DIAMOND for sequence alignment and OrthoLoger for inferring orthologs, to improve the accuracy of functional annotation. What sets Hayai-Annotation v3 apart is its ability to create network graphs and heatmaps linking orthologs with GO (molecular function and biological process), helping researchers infer the function of uncharacterized genes and visualize gene distribution across different species.

Through comparison, the authors found that Hayai-Annotation v3 performs comparably to the standard tool BLAST2GO in terms of accuracy. Additionally, it provides network analysis of orthologs in an unprecedented way by associating orthologs with GO, allowing for a comprehensive understanding of gene distribution and function across species. In this paper, the authors evaluated the performance of Hayai-Annotation v3 using peptide sequences from three plants (Arabidopsis, cultivated rice, and wild rice), and benchmarked it against other tools.

In conclusion, this paper aims to address the efficiency and accuracy issues in plant gene function annotation, providing a new perspective for studying plant gene evolution and adaptation through innovative network analysis methods.