dna-methylation-array-data-import

DNA Methylation Array Data Import

Summary

Load raw DNA methylation array data (IDAT files or beta-value matrices) from HumanMethylation450 or EPIC arrays into R, verifying correct probe counts before downstream filtering and analysis. This skill ensures data integrity at the entry point of the methylation analysis pipeline.

When to use

You have raw .idat files or beta-valued matrices from Illumina HumanMethylation450 (450K) or EPIC array experiments and need to import them into R for quality control and downstream analysis. Apply this skill at the very start of a methylation study before any probe filtering, normalization, or statistical testing.

When NOT to use

• Data has already been loaded and filtered; you are starting mid-pipeline with a processed feature table
• Working with non-Illumina methylation platforms (e.g., whole-genome bisulfite sequencing, enzymatic methyl-seq) — ChAMP is designed specifically for 450K and EPIC beadarray data
• Input is already a normalized or batch-corrected matrix; this skill addresses raw data import, not downstream corrections

Inputs

• .idat raw intensity files (paired Red and Green channel files per sample)
• Beta-value matrix (numeric matrix with CpG probes as rows, samples as columns)
• Sample metadata or phenotype file (optional but recommended for context)

Outputs

• ChAMP data object containing loaded probe intensities or beta values
• Pre-filter probe count (485,512 for 450K or 867,531 for EPIC)
• Sample-level quality metrics and import summary

How to apply

Use ChAMP's champ.load() or champ.import() functions to read data from .idat files or a beta-value matrix. For 450K arrays, verify the pre-filter probe count equals 485,512; for EPIC arrays, verify it equals 867,531. These counts confirm successful loading before quality-based filtering removes low-quality or cross-hybridizing probes. Extract the returned probe count from the output object and compare against the expected reference values for your array type. If counts deviate significantly, check for file corruption, incomplete sample sets, or array type mismatch.