sbov {TxRegInfra} | R Documentation |
prototype of subsetter for mongo resource
sbov(rme, gr, map = basicCfieldsMap(), docTypeName = "type", simplify = TRUE)
rme |
RaggedMongoExpt instance |
gr |
GRanges instance to subset by |
map |
list with one element per document type telling what fields are chr, start, stop |
docTypeName |
character(1) naming column of colData(rme) that has document type |
simplify |
logical(1) if TRUE, a GRanges is assembled; if FALSE, a RaggedExperiment is returned |
if simplify is TRUE (the default), a GRanges is returned with assay components in the mcols; otherwise a RaggedExperiment instance, with metadata component 'docType' set to propagate the input type
This is a placeholder for a proper subsetByOverlaps method. While it would appear sensible to allow subsetting for diverse assay types (e.g., eQTL, DGF), in version 1.5.x this is not supported. Each call can operate on only one sample type.
requireNamespace('mongolite') if (verifyHasMongoCmd()) { # for makeColData, which must be able to enumerate collections, # and thus must be able to run system (as opposed to mongolite function) 'mongo' m1 = mongolite::mongo(url=URL_txregInAWS(), db='txregnet') #cd = makeColData(url=URL_txregInAWS(), db='txregnet') cd = TxRegInfra::basicColData rme1 = RaggedMongoExpt(m1, cd[which(cd$type=='FP'),][seq_len(8),]) BiocParallel::register(BiocParallel::SerialParam()) qrng = GRanges('chr1', IRanges(1e6, 1.5e6)) GenomeInfoDb::genome(qrng) = "hg19" ss = sbov(rme1, GRanges('chr1', IRanges(1e6, 1.5e6)), simplify=FALSE) }