iRRBS

iRRBS corrects the 3' trimming of reduced representation bisulfite sequencing (RRBS) reads to prevent false positive differentially methylated site (DMS) identifications by identifying and hiding non-trimmed cytosines from methylation calling.

Key Features:

• Non-trimmed cytosine identification: Detects end-repaired cytosines at the 3' end of RRBS reads that Trim Galore fails to remove when the end-repair overlaps with the 3' end of sequencing reads.
• Methylation calling correction: Hides identified non-trimmed cytosines from methylation calling to eliminate bias in methylation level estimation.
• Low false positive rate: Achieves a maximum false positive rate of 0.5% in DMS identification.
• Cross-species and cross-laboratory compatibility: Validated on four datasets from four laboratories and two different species.

Scientific Applications:

• DNA methylation analysis: Enables accurate determination of DNA methylation at CpG-rich regions of the genome using RRBS data.
• Differential methylation detection: Prevents erroneous identification of up to >50% false positive differentially methylated sites (DMS) under certain conditions.
• RRBS pipeline integration: Implemented as a Python package for incorporation into standard RRBS computational analysis pipelines.

Methodology:

iRRBS analyzes RRBS reads after Trim Galore processing to identify non-trimmed 3' cytosines resulting from end-repair during MspI-digested DNA library preparation, then hides these cytosines from methylation calling.