Archival fixed tissues hold key insights into the evolutionary history of RNA viruses and the associated host immune response, yet access to the RNA sequence data is limited by a lack of robust methods for RNA extraction and sequence retrieval from these tissue types. Here we compared three commercial RNA extraction techniques (bead, column, and phase-based) on five fixed human brain tissues done in triplicate, that have been stored for up to 43 years. We found that for this sample set, bead-based extractions captured longer molecules and yielded a greater proportion of unique reads when aligned to the human genome, than did column and phase-based extraction methods. Via the incorporation of multiple extraction replicates, we quantified the variability in sequencing metrics resulting from tissue sample and extraction technique heterogeneity. Additionally, we compared pre- and post-sequencing metrics and found that the former poorly predicted post-sequencing on-target success. Our findings help inform future research on the recovery of RNA from archival fixed tissues.
Keywords: RNA; archival; extraction; fixed tissue; museum collections.
We performed RNA extraction using three commercial kits based on different underlying methodologies (bead, column, and phase-based) from five FFPE brain tissues up to 43 years old. Pre-sequencing metrics such as RIN, RNA concentration, and qPCR cycle threshold were measured alongside post-sequencing metrics such as insert length distribution, library complexity, and human genome and transcriptome mapping.