Apollo: Collaborative and Scalable Manual Genome Annotation

Manual curation is crucial to improving genome annotation quality. However, it is impossible for a single individual to fully curate a genome with precise biological fidelity. The traditional centralized model of genome annotation has given way to a geographically dispersed community annotation model. Breaking down large amounts of genomic data into manageable portions and mobilizing groups of researchers to extract an accurate representation of the biology from these data distills invaluable knowledge.

Apollo offers a web-based environment for distributed collaborators to manually revise, edit, and share annotations in real-time (akin to Google Docs). It provides a fast interface for visualizing genomic data and allows extensive changes to be rapidly made without server round-trips. Gene models, protein alignments, and expression and variant data inform structural annotations. Users may add metadata, including references to other databases, functional assignments to genes and gene products using Gene Ontology (GO) terms, and references to literature in support of these annotations. Apollo can easily integrate into workflows via a suite of web services, which has facilitated integration within the Galaxy platform.

Current work: 1) visualization and annotation of variant data; 2) transformation of genomic coordinates into synthetic regions, allowing folding (hiding from view) of intra- and intergenic regions of the genome as well as the ability to combine scaffolds. This facilitates annotation of genomic elements split across two or more regions of the assembly. We will describe Apollo’s functionality, briefly introduce its architecture, and offer details of future plans. Learn more at http://genomearchitect.org/.

https://github.com/GMOD/Apollo