WhatsGNU: a tool for identifying proteomic novelty
Year: 2019
Type: other
Abstract: For many bacterial species, thousands of whole genome sequences are now available in public databases. To understand diversity and novelty in such enormous collections we need computationally scalable tools that can quickly contextualize genomes based on their similarities to known variation and identify features of each genome that make them unique. Here we present a tool based on exact match proteomic compression that, in seconds, delivers a whole proteome report classifying any new genome to the strain level, and provides a detailed report of protein variants that may have novel functional differences. The technique utilizes the natural variation in public databases to rank protein sequences based on the number of observed exact protein matches (the Gene Novelty Unit (GNU) score) in all known genomes. The GNU score affords a convenient way to measure known protein diversity, describe protein conservation, identify the closest match genomes, and, most importantly, assay for novelty. We use this technique to characterize the panallelome of Mycobacterium tuberculosis , Pseudomonas aeruginosa and Staphylococcus aureus , and we provide a tool, WhatsGNU, that can be used to quickly create whole protein reports and classify genomes. We suggest that this technique could be extended to most bacterial species for which large numbers of genomes are currently available. WhatsGNU is available from https://github.com/ahmedmagds/WhatsGNU .
Authors: Ahmed Magdy Moustafa, Paul Planet
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: gold
OpenAlex ID: W3161235967