Background: Co-sequencing of messenger ribonucleic acid (mRNA) and micro ribonucleic acid (miRNA) across a time series (1, 3, 6, 24, and 48 h post injury) was used to identify potential miRNA-gene interactions during pancreatic injury, associate serum and tissue levels of candidate miRNA biomarkers of pancreatic injury, and functionally link these candidate miRNA biomarkers to observed histopathology. RNAs were derived from pancreatic tissues obtained in experiments characterizing the serum levels of candidate miRNA biomarkers in response to acute pancreatic injury in rats.
Results: No correlation was discovered between tissue and serum levels of the miRNAs. A combination of differential gene expression, novel delayed anti-correlation analysis and experimental database interrogation was used to identify messenger RNAs and miRNAs that experienced significant expression change across the time series, that were negatively correlated, that were complementary in sequence, and that had experimentally supported relationships. This approach yielded a complex signaling network for future investigation and a link for the specific candidate miRNA biomarkers, miR-216a-5p and miR-217-5p, to cellular processes that were in fact the prominent histopathology observations in the same experimental samples. RNA quality bias by treatment was observed in the study samples and a statistical correction was applied. The relevance and impact of that correction on significant results is discussed.
Conclusion: The described approach allowed extraction of miRNA function from genomic data and defined a mechanistic anchor for these miRNAs as biomarkers. Functional and mechanistic conclusions are supported by histopathology findings.
Keywords: Biomarkers; Co-sequencing; Delayed anti-correlation; Pancreatic injury; mRNA; miRNA.