Changes in version 1.21.5 o Added functionality to the gene set testing functions to allow the user to limit the input CpGs to genomic regions of interest with the genomic.features parameter. o Added sig.genes parameter to the gene set testing functions to allow the user to add the significant genes overlapping each gene set to the output. o Added a fix to gsameth to avoid p-values of exactly zero being reported. o Modified the default for coef in the varFit function so the default is all columns of the design matrix used to compute Levene Residuals. o Updated vignette to include examples of how to perform gene set testing following a region level analysis Changes in version 1.21.4 o Bug fix: Fixed bug that resulted in negative value for odds. Bug only occurred in rare cases when CpG universe was smaller than all CpGs on array AND if some large collections had more genes than the universe. Changes in version 1.20.3 o Added fract.counts logical parameter to enable fractional counting of CpGs to be turned off in gometh, gsameth, goregion and gsaregion. This switches off the adjustment for CpGs that map to multiple genes. This is primarily intended for comparison with earlier versions and we strongly encourage it to be set to TRUE. Changes in version 1.20.0 o Updated gsameth to take into account that some CpGs map to multiple genes. For a small number of gene families, this previously caused their associated GO categories/gene sets to be erroneously overrepresented and thus highly significant. o Updated gometh to call gsameth to perform statistical testing for GO categories and KEGG pathways; goana and kegga are no longer used. o Replaced use of the AnnotationDbi 'toTable' function with 'select' to speed up execution. o The 'topGSA' function is now used to look at the top ranked results from both gometh and gsameth. The vignette has been modified accordingly. Changes in version 1.15.2 o Updated getMappedEntrezIDs, gometh and gsameth to to speed up execution by taking the array annotation in as an optional argument. o missMethyl now uses the latest IlluminaHumanMethylationEPICanno.ilm10b2.hg19 annotation by default for EPIC arrays. Changes in version 1.15.1 o Added getAdjusted function for extracting RUVm adjusted data for visualisation purposes o Updated vignette to demonstrate use of getAdjusted function o Vignette now includes an example of how to handle cases with RUVm where number of samples is greater than number of Illumina negative controls Changes in version 1.7.3 o modified gene set testing functions gometh and gsameth to accommodate EPIC arrays Changes in version 1.5.1 o New gene set testing function gsameth() added, kegg testing functionality added to gometh. Changes in version 1.1.10 o This version of the package contains functions to perform SWAN normalisation, RUV normalisation and differential methylation analysis, differential variability analysis and gene ontology testing for the 450K array. Changes in version 1.1.3 o Added the gometh function to perform gene ontology analysis. Changes in version 1.1.2 o Added functions to perform RUV normalization and differential methylation analysis. Changes in version 0.99.9 o This version of the package has all calculations in matrix formulation and handles nuisance parameters appropriately. Changes in version 0.99.7 o Added new function, contrasts.varFit. Changes in version 0.99.6 o Modified getLeveneResiduals Changes in version 0.99.5 o Modified the varFit function to accept DGEList objects for differential variability testing for RNA-Seq data. Changes in version 0.99.4 o First version of package contains functions to perform SWAN normalisation and differential variability analysis for DNA methylation data from Illumina's Infinium HumanMethylation450 beadchip.