grep not giving specific results in R-CodePudding

I want to filter paths matching only some values ( "chr1" "chr11" "chr16" "chr17" "chr2" "chr5" "chr6" "chr7") in a list of paths. However my results includes additional chr#

This is the items i want to filter

> sort(chrm_to_filter$chr)
 "chr1"  "chr11" "chr16" "chr17" "chr2"  "chr5"  "chr6"  "chr7"

My data looks something like this

print(path_per_chr_tabix)
"/path_to_file/merged_modified_per_base_calling.chr1.bgz" 
"/path_to_file/merged_modified_per_base_calling.chr10.bgz"
"/path_to_file/merged_modified_per_base_calling.chr11.bgz"
"/path_to_file/merged_modified_per_base_calling.chr12.bgz"
"/path_to_file/merged_modified_per_base_calling.chr13.bgz"
"/path_to_file/merged_modified_per_base_calling.chr14.bgz"
"/path_to_file/merged_modified_per_base_calling.chr15.bgz"
"/path_to_file/merged_modified_per_base_calling.chr16.bgz"
"/path_to_file/merged_modified_per_base_calling.chr17.bgz"
"/path_to_file/merged_modified_per_base_calling.chr18.bgz"
"/path_to_file/merged_modified_per_base_calling.chr19.bgz"
"/path_to_file/merged_modified_per_base_calling.chr2.bgz" 
"/path_to_file/merged_modified_per_base_calling.chr3.bgz" 
"/path_to_file/merged_modified_per_base_calling.chr4.bgz" 
"/path_to_file/merged_modified_per_base_calling.chr5.bgz" 
"/path_to_file/merged_modified_per_base_calling.chr6.bgz" 
"/path_to_file/merged_modified_per_base_calling.chr7.bgz" 
"/path_to_file/merged_modified_per_base_calling.chr8.bgz" 
"/path_to_file/merged_modified_per_base_calling.chr9.bgz" 
"/path_to_file/merged_modified_per_base_calling.chrm.bgz" 
"/path_to_file/merged_modified_per_base_calling.chrX.bgz" 
"/path_to_file/merged_modified_per_base_calling.chrY.bgz"

#find which data to load to save memory

subset_tabix_paths_to_load <- path_per_chr_tabix[
    grep(
        paste0(sort(chrm_to_filter$chr), collapse="|"), 
        path_per_chr_tabix)
    ]
message("these are the files we will be workign with for now- ")
print(subset_tabix_paths_to_load)

"/paths/merged_modified_per_base_calling.chr1.bgz" 
"/paths/merged_modified_per_base_calling.chr10.bgz"
"/paths/merged_modified_per_base_calling.chr11.bgz"
"/paths/merged_modified_per_base_calling.chr12.bgz"
"/paths/merged_modified_per_base_calling.chr13.bgz"
"/paths/merged_modified_per_base_calling.chr14.bgz"
"/paths/merged_modified_per_base_calling.chr15.bgz"
"/paths/merged_modified_per_base_calling.chr16.bgz"
"/paths/merged_modified_per_base_calling.chr17.bgz"
"/paths/merged_modified_per_base_calling.chr18.bgz"
"/paths/merged_modified_per_base_calling.chr19.bgz"
"/paths/merged_modified_per_base_calling.chr2.bgz" 
"/paths/merged_modified_per_base_calling.chr5.bgz" 
"/paths/merged_modified_per_base_calling.chr6.bgz" 
"/paths/merged_modified_per_base_calling.chr7.bgz"

CodePudding user response：

The problem is that one of your search strings is 'chr1', which is found in 'chr1', 'chr10', 'chr11', etc. If you change your search strings to 'chr1.bgz' etc, it should work.

CodePudding user response：

Here's an approach using sub which

> filter_chr <- c("chr1",  "chr11", "chr16", "chr17", "chr2",  "chr5",  "chr6",  "chr7") 
> string[which(sub(".*\\.(chr\\d )\\..*$", "\\1", string) %in% filter_chr)]
[1] "/path_to_file/merged_modified_per_base_calling.chr1.bgz" 
[2] "/path_to_file/merged_modified_per_base_calling.chr11.bgz"
[3] "/path_to_file/merged_modified_per_base_calling.chr16.bgz"
[4] "/path_to_file/merged_modified_per_base_calling.chr17.bgz"
[5] "/path_to_file/merged_modified_per_base_calling.chr2.bgz" 
[6] "/path_to_file/merged_modified_per_base_calling.chr5.bgz" 
[7] "/path_to_file/merged_modified_per_base_calling.chr6.bgz" 
[8] "/path_to_file/merged_modified_per_base_calling.chr7.bgz"

string has the following structure:

c("/path_to_file/merged_modified_per_base_calling.chr1.bgz", 
"/path_to_file/merged_modified_per_base_calling.chr10.bgz", "/path_to_file/merged_modified_per_base_calling.chr11.bgz", 
"/path_to_file/merged_modified_per_base_calling.chr12.bgz", "/path_to_file/merged_modified_per_base_calling.chr13.bgz", 
"/path_to_file/merged_modified_per_base_calling.chr14.bgz", "/path_to_file/merged_modified_per_base_calling.chr15.bgz", 
"/path_to_file/merged_modified_per_base_calling.chr16.bgz", "/path_to_file/merged_modified_per_base_calling.chr17.bgz", 
"/path_to_file/merged_modified_per_base_calling.chr18.bgz", "/path_to_file/merged_modified_per_base_calling.chr19.bgz", 
"/path_to_file/merged_modified_per_base_calling.chr2.bgz", "/path_to_file/merged_modified_per_base_calling.chr3.bgz", 
"/path_to_file/merged_modified_per_base_calling.chr4.bgz", "/path_to_file/merged_modified_per_base_calling.chr5.bgz", 
"/path_to_file/merged_modified_per_base_calling.chr6.bgz", "/path_to_file/merged_modified_per_base_calling.chr7.bgz", 
"/path_to_file/merged_modified_per_base_calling.chr8.bgz", "/path_to_file/merged_modified_per_base_calling.chr9.bgz", 
"/path_to_file/merged_modified_per_base_calling.chrm.bgz", "/path_to_file/merged_modified_per_base_calling.chrX.bgz", 
"/path_to_file/merged_modified_per_base_calling.chrY.bgz")

CodePudding user response：

lapply(sort(chrm_to_filter$chr), 
      function(chr) {
          path_per_chr_tabix[grep(paste0(chr,"."),
                     path_per_chr_tabix, 
                     fixed = TRUE)]
       })|> 
      unlist()

#[1] "/path_to_file/merged_modified_per_base_calling.chr1.bgz" 
#[2] "/path_to_file/merged_modified_per_base_calling.chr11.bgz"
#[3] "/path_to_file/merged_modified_per_base_calling.chr16.bgz"
#[4] "/path_to_file/merged_modified_per_base_calling.chr17.bgz"
#[5] "/path_to_file/merged_modified_per_base_calling.chr2.bgz" 
#[6] "/path_to_file/merged_modified_per_base_calling.chr5.bgz" 
#[7] "/path_to_file/merged_modified_per_base_calling.chr6.bgz" 
#[8] "/path_to_file/merged_modified_per_base_calling.chr7.bgz"