Seqpac: a framework for sRNA-seq analysis in R using sequence-based counts

Area of Science:

• Bioinformatics
• Genomics
• Molecular Biology

Background:

• Feature-based counting in RNA sequencing (RNA-seq) aligns sequences to predefined features, potentially losing critical information and traceability.
• Small RNA sequencing (sRNA-seq) data is particularly susceptible to misinterpretation due to sequence diversity and multi-mapping.
• The 'traceability dilemma' arises from alignment-based strategies that compromise data lineage.

Purpose of the Study:

• To introduce Seqpac, an R package designed for small RNA sequencing (sRNA-seq) analysis.
• To present a novel strategy that preserves raw sequence integrity for enhanced data traceability in sRNA-seq.
• To demonstrate Seqpac's ability to uncover biases and provide new insights in previously analyzed sRNA-seq datasets.

Main Methods:

• Development of Seqpac, an R package implementing a flexible framework for sRNA-seq analysis.
• Preservation of read sequence integrity throughout the analysis pipeline.
• Application of Seqpac to published biological datasets for comparative analysis.

Main Results:

• Seqpac successfully preserves sequence integrity, ensuring full data lineage traceability.
• Analysis using Seqpac revealed hidden biases in previously published sRNA-seq studies.
• The package provided novel biological insights not apparent with traditional feature-based counting methods.

Conclusions:

• Seqpac offers a superior approach to sRNA-seq analysis by maintaining data integrity and traceability.
• This strategy enhances the interpretability of sRNA-seq data, especially for complex datasets.
• The principles behind Seqpac could be applied to other transcriptomic workflows to improve reproducibility and address the replication crisis.