Comprehensive network prediction for any fully sequence genome in STRING database
Zurich Seminars in Bioinformatics - Damian Szklarczyk
Abstract Every day new genomes are sequenced and existing genomes are re-sequenced and re-annotated. In the new version of STRING the user can submit any fully sequenced genome for complete network and functional annotation. To do so. will require only minimal input from the user in a form of the proteome in a FASTA format and, if known, a taxonomical clade of the given genome.
Utilizing hierarchical orthology relationships from the eggNOG database precomputed for more than 1600 taxonomical clades the STRING algorithm chooses the most precise orthologous group for each protein and leverages all the data already included in the database for this group and its parent groups to predict both the functional neighbourhood and the functional annotation of each protein in the user input. The resulting dataset is explorable in the same manner as any other organism present in the STRING database: it’s sharable, with fully working user interface, extensive REST API access, a set of enrichment analysis tools and clustering methods, and simple flat-files available for download.
- 12:15 UZH Irchel Y55-l-06/08 and ZOOM Call