A modular class-aware workflow for small RNA sequencing analysis using mouse sperm as a case study.

BACKGROUND

Small RNA sequencing analysis is challenging because RNA classes differ in biogenesis, sequence redundancy, genomic organization, and annotation reliability. Integrated workflows accommodating these constraints remain limited, particularly for fragment-level and cluster-level analysis.

METHODS

We present a reproducible, containerized, class-aware workflow for small RNA sequencing analysis, using mouse sperm as a case study. The workflow combines standardized preprocessing with complementary annotation and quantification strategies for microRNAs (miRNAs), transfer RNA-derived small RNAs (tsRNAs), ribosomal RNA-derived small RNAs (rsRNAs), and PIWI-interacting RNA (piRNA)-enriched genomic clusters. Using sperm small RNA data from offspring of lipopolysaccharide (LPS)-exposed male mice, we compared integrated-reference mapping, multi-class annotation, fragment-level tsRNA profiling, and genome-based piRNA cluster analysis, with custom modules for locus-aware harmonization and condition-specific cluster analysis.

RESULTS

Integrated-reference mapping aligned 88.17% of reads and retained 690 features after filtering. It identified 11 differentially expressed miRNAs between LPS and controls, while other classes showed limited signal. Fragment-level profiling improved tsRNA resolution. piRNA cluster analysis identified 958 control and 940 LPS clusters, with 18 control-specific and no LPS-specific clusters.

CONCLUSION

This workflow supports transparent, reproducible, class-aware interpretation of small RNA sequencing data while emphasizing cautious interpretation of piRNA-enriched signals from total small RNA sequencing.