Interpretation of biological experiments changes with evolution of the Gene Ontology and its annotations

Abstract

Gene Ontology (GO) enrichment analysis is ubiquitously used for interpreting high throughput molecular data and generating hypotheses about underlying biological phenomena of experiments. However, the two building blocks of this analysis — the ontology and the annotations — evolve rapidly. We used gene signatures derived from 104 disease analyses to systematically evaluate how enrichment analysis results were affected by evolution of the GO over a decade. We found low consistency between enrichment analyses results obtained with early and more recent GO versions. Furthermore, there continues to be a strong annotation bias in the GO annotations where 58% of the annotations are for 16% of the human genes. Our analysis suggests that GO evolution may have affected the interpretation and possibly reproducibility of experiments over time. Hence, researchers must exercise caution when interpreting GO enrichment analyses and should reexamine previous analyses with the most recent GO version.

GSAn: an alternative to enrichment analysis for annotating gene sets

“…That problem can be solved using classical enrichment methods, such as DAVID or g:Profiler. However, these methods focus on the most studied genes that may provide annotations covering a limited number of annotated genes [8,6,7]. Another problem is the redundant information within annotations that may increase the difficulty in interpreting results when no a posteriori analysis is performed.…”

“…The results are thus given as lists of related terms and an additional manual expertise is still required to synthesize the information. Moreover, significant limitations of enrichment-based methods have recently been reported [6,7].…”

“…First, these methods tend to focus on the most studied genes and provide gene set annotation results that cover a limited number of annotated genes [8,6,7].…”

Monitoring changes in the Gene Ontology and their impact on genomic data analysis

“…However, the effects of changes are not likely to be uniform across datasets nor easily predictable. Indeed, previous studies have been either anecdotal (considering a single or just a few examples [ 8 – 11 ]), with the largest study analyzing around 100 [ 12 ], or yielded mixed findings. Groß et al (2012) found that enrichment results were stable based on analysis of two hit lists.…”

“…Alam-Faruqe et al considered changes in results to be improvements due to focused curation, based on analysis of two datasets. Others have emphasized instability [ 11 , 12 ] or reported mixed impacts [ 9 ]. Given this variety of results and interpretations, there is clearly a need for researchers to assess the stability of their own specific enrichment results.…”