Abstract
The Gene Ontology (GO) is one of the most widely used resources in molecular and cellular biology, largely through the use of “enrichment analysis”. Despite the high importance of GO, there has been limited appreciation that it changes over time and limited evaluation of how these changes impact research. To help close this gap, we present GOTrack (https://gotrack.msl.ubc.ca), a web-based system and database that provides access to historical records and trends in the Gene Ontology and GO annotations (GOA). GOTrack gives users access to gene- and term-level information on annotations for nine model organisms (including human) as well as an interactive tool that measures the stability of enrichment results over time for user-provided “hit lists” of genes. To document the effects of GO/GOA evolution on enrichment analysis, we analyzed over 2500 previously published hit lists of human genes. We compared the results that would have been obtained around the time of publication to those obtained for a current edition of GO/GOA, calibrated using a null based on randomly paired hit lists. Overall, 53% of hit lists were considered to yield significantly stable enrichment results, despite 90% of them being more than 9.2 years old. Because stability is far from assured for any individual hit list, GOTrack can lead to more informed and cautious application of GO to genomics research.
