MethylToSNP

MethylToSNP identifies single nucleotide polymorphisms (SNPs) within Illumina DNA methylation array data to distinguish genetic variation from methylation signals altered by sodium bisulfite conversion.

Key Features:

• Algorithmic Detection of SNP Patterns: Detects characteristic data patterns in Illumina methylation array measurements that indicate confounding SNPs at probe sites.
• Reliability Scoring System: Assigns a reliability score to SNP predictions enabling threshold-based filtering.
• Calibration on Diverse Data Sets: Parameters and thresholds are calibrated using simulated and real methylation datasets.
• Validation through Multiple Methods: Predictions are validated against SNP databases and by bisulfite and genomic sequencing.

Scientific Applications:

• Minimizing Data Loss: Identifies SNPs specific to individual samples to avoid blanket exclusion of problematic probes and reduce data loss.
• Discovery of Novel Polymorphisms: Enables detection of previously uncharacterized polymorphisms in populations with limited genetic reference data.
• Functional Genomic Insights: Detects variants within functional regions such as CTCF sites and enhancers to inform regulatory analyses.

Methodology:

Analyzes Illumina methylation array data (450K, 850K EPIC, 27K) from populations including Yoruba in Ibadan (YRI), CEPH (European descent), and KhoeSan; calibrates parameters on simulated and real methylation datasets; and validates predictions against SNP databases and by bisulfite and genomic sequencing.