Difficulty filtering non-standard chromosomes

[dustin-mullaney]

Hello,
I am attempting to use the LinkPeaks() function, but I am having trouble getting it to work. I am working with 10x atac data that was aligned with hg38. I use the following code to create the chromatin assay:

annotation <- GetGRangesFromEnsDb(ensdb = EnsDb.Hsapiens.v86)
seqlevelsStyle(annotation) <- "UCSC"
obj[["ATAC"]] <- CreateChromatinAssay(
    counts = atac_counts,
    sep = c(":", "-"),
    fragments = frag.path,
    annotation = annotation
)

I did not use Macs2 to call the peaks .

When I run RegionStats() before LinkPeaks() (after all the standard data processing) I get the following error:

>   obj <- RegionStats(obj, genome = BSgenome.Hsapiens.UCSC.hg38)
Error in .getOneSeqFromBSgenomeMultipleSequences(x, name, start, NA, width,  : 
  sequence KI270728.1 not found
In addition: Warning message:
In .Seqinfo.mergexy(x, y) :
  Each of the 2 combined objects has sequence levels not in the other:
  - in 'x': chrM, chr1_GL383518v1_alt, chr1_GL383519v1_alt, chr1_GL383520v2_alt, chr1_KI270759v1_alt, chr1_KI270760v1_alt, chr1_KI270761v1_alt, chr1_KI270762v1_alt, chr1_KI270763v1_alt, chr1_KI270764v1_alt, chr1_KI270765v1_alt, chr1_KI270766v1_alt, chr1_KI270892v1_alt, chr2_GL383521v1_alt, chr2_GL383522v1_alt, chr2_GL582966v2_alt, chr2_KI270767v1_alt, chr2_KI270768v1_alt, chr2_KI270769v1_alt, chr2_KI270770v1_alt, chr2_KI270771v1_alt, chr2_KI270772v1_alt, chr2_KI270773v1_alt, chr2_KI270774v1_alt, chr2_KI270775v1_alt, chr2_KI270776v1_alt, chr2_KI270893v1_alt, chr2_KI270894v1_alt, chr3_GL383526v1_alt, chr3_JH636055v2_alt, chr3_KI270777v1_alt, chr3_KI270778v1_alt, chr3_KI270779v1_alt, chr3_KI270780v1_alt, chr3_KI270781v1_alt, chr3_KI270782v1_alt, chr3_KI270783v1_alt, chr3_KI270784v1_alt, chr3_KI270895v1_alt, chr3_KI270924v1_alt, chr3_KI270934v1_alt, chr3_KI270935v1_alt, chr3_KI270936v1_alt, chr3_KI270937v1_alt, chr4_GL000 [... truncated]

Going off of this post, I assume I need to use keepStandardChromosomes() to filter the peaks in non-standard chromosomes. However what I have tried so far hasn't solved my issue. I tried running:

annotation <- GetGRangesFromEnsDb(ensdb = EnsDb.Hsapiens.v86)
seqlevelsStyle(annotation) <- "UCSC"

keepStandardChromosomes(annotation)

obj[["ATAC"]] <- CreateChromatinAssay(
    counts = atac_counts,
    sep = c(":", "-"),
    fragments = frag.path,
    annotation = annotation
)

Should I instead be using keepStandardChromosomes() elsewhere in the code?

Any suggestions are appreciated!

[vgilbart]

In case someone else stumbles on this question, here's how I filtered out peaks on non-standard chromosomes in Seurat object:

main.chroms <- standardChromosomes(BSgenome.Mmusculus.UCSC.mm10)
keep.peaks <- rownames(seurat)[lapply(strsplit(rownames(seurat), '-'), `[[`, 1) %in% main.chroms]
seurat <- subset(seurat, features = keep.peaks)

And how to filter the genome on non-standard chromosomes:

# from https://support.bioconductor.org/p/83588/
keepBSgenomeSequences <- function(genome, seqnames)
{
    stopifnot(all(seqnames %in% seqnames(genome)))
    genome@user_seqnames <- setNames(seqnames, seqnames)
    genome@seqinfo <- genome@seqinfo[seqnames]
    genome
}
# main.chroms <- standardChromosomes(BSgenome.Mmusculus.UCSC.mm10)
filtered_genome <- keepBSgenomeSequences(BSgenome.Mmusculus.UCSC.mm10, main.chroms)