RRBS feature annotation

RRBS feature annotation

Format

A data frame with 7585076 rows and 12 variables:

Chr

integer, chromosome

Locus

character, base pair range of feature

EntrezID

character, Entrez ID of closest gene

Symbol

character, gene symbol of closest gene

Strand

character, strand

Width

integer, width of the genomic locus represented by the feature

NumSites

integer, the number of sites merged to generate the feature, sites are merged by their correlation pattern in the data using an unsupervised analysis

Sites

character, a comma separated string with the specific sites that were merged to generate the feature

LocStart

integer, feature start in base pairs

LocEnd

integer, feature end in base pairs

tissue

character, tissue abbreviation, one of TISSUE_ABBREV

feature_ID

character, MoTrPAC feature identifier

Details

METHYL feature annotation is only available via download from Google Cloud Storage: https://storage.googleapis.com/motrpac-rat-training-6mo-extdata/epigen-rda/METHYL_FEATURE_ANNOT.rda. You can use MotrpacRatTraining6mo::load_methyl_feature_annotation() to download and return this file.

Only CpG sites with methylation coverage of >=10 in all samples were included for downstream analysis, and normalization was performed separately in each tissue. Individual CpG sites were divided into 500 base-pair windows and were clustered using the Markov Clustering algorithm via the MCL R package (Jager, 2015). To apply MCL, for each 500 base-pair window an undirected graph was constructed, linking individual sites if their correlation was >=0.7. MCL was chosen for this task as it: (1) determines the number of clusters internally, (2) identifies homogeneous clusters, and (3) keeps single sites that are not correlated with either sites as singletons (clusters of size one).

Given these sites, this table was generated using MotrpacRatTraining6mo::get_peak_annotations(). relationship_to_gene is the shortest distance between the feature and the start or end of the closest gene. It is 0 if the feature has any overlap with the gene. custom_annotation fixes many issues with the ChIPseeker annotation (v1.22.1).