The core workflow of GUNC consists of three modules (see Fig. Fig.1c).1c). First, for any query genome, genes are called using prodigal [44], although per-gene protein sequences can alternatively be supplied by the user directly. Protein sequences are then mapped against representative genomes in the GUNC database (derived from species-representative genomes in proGenomes 2.1 [34]) using diamond [45], retaining best hits (-k 1) without applying an evalue filter (-e 1) as alternative filtering is applied downstream. Annotated plasmids and other non-chromosomal genomic elements are excluded from the reference to reduce nonspecific hits between lineages within plasmid host range. Moreover, the reference set was semi-manually curated, removing clear cases of genomic chimerism.

For each query gene, taxonomic annotations at 7 levels (kingdom, phylum, class, order, family, genus, species) are inherited from the best hit via the manually curated proGenomes 2.1 taxonomy. To filter against mapping noise, taxonomic clade labels recruiting less than 2% of all mapped genes are dropped. GUNC scores (see below) are then calculated based on inferred taxonomic labels, query gene contig membership, sequence identity to database hits, and the fraction of mapped and filtered hits. Finally, GUNC offers a visualization module to automatically generate interactive Sankey alluvial diagrams of contig-level taxonomic annotations to enable manual curation and exploration of flagged genomes.

GUNC is implemented in Python3, all code is open source and available at https://github.com/grp-bork/gunc and through bioconda [46] under a GPLv3+ license. Based on database size and resource requirements, GUNC can be run locally on a personal computer but is also highly parallelizable in a cluster environment.

Note: The content above has been extracted from a research article, so it may not display correctly.



Q&A
Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.



We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.