Oral Presentation Australasian RNA Biology and Biotechnology Association 2024 Conference

High-Accuracy RNA DEGRADATION ESTIMATION for Unbiased Transcriptome Comparisons with INDEGRA (108987)

Alice Cleynen 1 , Agin Ravindran 1 , Aditya Sethi 1 , Bhavika Kumar1 1 , Katrina Woodward 1 , Robert Weatheritt 2 , Stéphane Robin 3 , Eduardo Eyras 1 , Nikolay Shirokikh 1
  1. JCSMR, Canberra, ACT, Australia
  2. Tampere University, Tampere, Finland
  3. Sorbone Université, Paris, France

Background: RNA carries immense amounts of information about transcription, splicing, translation and epitranscriptome layers of gene control. However, RNA is also a subject of rapid chemical and enzymatic decay (degradation) due to native metabolism and, commonly, sample handling. Variability in the RNA integrity creates a source of disparity between the compared samples, presenting a major challenge in RNA biology.

 

Aims and Methods: Taking advantage of the absence of amplification bias, minimal artifacts induced by library construction and access to the full-length RNA molecules of Direct RNA Sequencing (DRS), we provide a universal and robust transcriptome-wide RNA integrity measure, the Direct Transcriptome Integrity number (DTI).

We extensively validate the DTI model, including through controlled random chemical fragmentation of RNA. Using DTI in an integrated software pipeline for Integrity and Degradation of RNA Assessment (INDEGRA), we demonstrate that we can isolate biological RNA degradation from technical sample integrity and link the apparent degradation profile with the underlying fragmentation rate of the RNA.

Applying INDEGRA to a large DRS dataset of 6 tissues across 6 species, we show that DTI provides additional information to the standard differential expression, exposing transcript-specific RNA degradation patterns. We identify differential biological degradation even across complicated samples of uneven technical integrity. Furthermore, INDEGRA corrects degradation bias in expression quantification, and is directly pluggable into popular differential expression tools such as DESeq2, edgeR or limma-voom, reducing degradation-invoked false hits in studies of any number of samples and conditions.

 

Conclusion: INDEGRA for the first time allows the comprehensive analysis of RNA degradation, providing an interpretable sample integrity metric together with per-transcript degradation quantification, and a differential degradation pipeline to compare RNA decay in several conditions. INDEGRA broadly enables new applications in dissecting RNA degradation through standard DRS in any setting.