Improving gene function predictions using independent transcriptional components

The interpretation of high throughput sequencing data is limited by our incomplete functional understanding of coding and non-coding transcripts. Reliably predicting the function of such transcripts can overcome this limitation. Here we report the use of a consensus independent component analysis and guilt-by-association approach to predict over 23,000 functional groups comprised of over 55,000 coding and non-coding transcripts using publicly available transcriptomic profiles. We show that, compared to using Principal Component Analysis, Independent Component Analysis-derived transcriptional components enable more confident functionality predictions, improve predictions when new members are added to the gene sets, and are less affected by gene multi-functionality. Predictions generated using human or mouse transcriptomic data are made available for exploration in a publicly available web portal. Our understanding of the function of many transcripts is still incomplete, limiting the interpretability of transcriptomic data. Here the authors use consensus-independent component analysis, together with a guilt-by-association approach, to improve the prediction of gene function.

Automated summary

The study introduces a consensus independent component analysis and guilt-by-association approach to predict gene function, aiming to overcome the limitation in interpreting transcriptomic data. The results demonstrate that transcriptional components derived from independent component analysis enable more confident functionality predictions, improve predictions with the addition of new members to gene sets, and are less affected by gene multi-functionality. Predictions based on human or mouse transcriptomic data are publicly available for exploration on a web portal.