Protein family neighborhood analyzer—ProFaNA

PeerJ

article

Bartosz Baranowski¹  Krzysztof Pawłowski¹ 

[1] Department of Biochemistry and Microbiology, Warsaw University of Life Sciences;Institute of Biochemistry and Biophysics, Polish Academy of Sciences;Department of Molecular Biology, University of Texas Southwestern Medical Center;Department of Translational Sciences, Lund University

关键词: Gene function prediction;    Genomic neighborhoods;    Protein domains;    Comparative genomics;   

DOI  :  10.7717/peerj.15715

学科分类：社会科学、人文和艺术（综合）

来源: Inra

【 摘 要 】

Background Functionally related genes are well known to be often grouped in close vicinity in the genomes, particularly in prokaryotes. Notwithstanding the diverse evolutionary mechanisms leading to this phenomenon, it can be used to predict functions of uncharacterized genes. Methods Here, we provide a simple but robust statistical approach that leverages the vast amounts of genomic data available today. Considering a protein domain as a functional unit, one can explore other functional units (domains) that significantly often occur within the genomic neighborhoods of the queried domain. This analysis can be performed across different taxonomic levels. Provisions can also be made to correct for the uneven sampling of the taxonomic space by genomic sequencing projects that often focus on large numbers of very closely related strains, e.g., pathogenic ones. To this end, an optional procedure for averaging occurrences within subtaxa is available. Results Several examples show this approach can provide useful functional predictions for uncharacterized gene families, and how to combine this information with other approaches. The method is made available as a web server at http://bioinfo.sggw.edu.pl/neighborhood_analysis.

【 授权许可】

CC BY