title: "HarmanData: example and reference data to use with batch correction package Harman"author: "Jason Ross and Yalchin Oytam"date: "r doc_date()"package: "r pkg_ver('HarmanData')"output:BiocStyle::html_documentvignette: >%\VignetteIndexEntry | Overview | Example gene expression data | IMR90 | NPM | OLF | Usage | Methylation reference data | Infinium5 | episcope | References

Introduction | Transcriptome data examples | Working with a simple transciptome microarray dataset | Running Harman | Inspecting results | Reconstruct the corrected data | How does the new data differ from the old data? | Another simple transciptome dataset | Evidence of batch effects | Working with very small datasets | More aggressive settings | Harman plots | Working with unbalanced and confounded data | Example dataset | Batch structure | Limma analysis | Methylation data examples | Loading 450K data | Appropriate normalisation | Harman correction of M | Clustering of methylation values | Implications for EWAS | Loading the reference matrix | Discovering clusters de novo in large EWAS studies | Thresholding | On de novo clustered data | Using a reference | Mass Spectrophotometry data example | Loading the mass-spec data | Preprocessing | Comparison to ComBat | IMR90 example data | Applying Harman and ComBat to adjust for known batches | Compare | Concluding remarks