CODEWITHSUNDEEP
rdataframesubset
user1636450
user1636450

Reputation: 13

Trouble with subset() in R returning incorrect values

I am attempting to subset a region of genetic data by its chromosome and position information. Unfortunately, my results are not within my parameters. Any help would be appreciated.

Here is my code:

    subset.by.region<- function(df,region.info, expansion=0){
     MBstart = as.numeric(region.info[[3]]) - expansion
     MBend = as.numeric(region.info[[4]]) + expansion
     chrom =as.numeric(region.info[[2]])
     print(chrom)
     print(MBstart)
     print(MBend)
     BPstart <- MBstart  * 1e6
     BPend <- MBend  * 1e6
     sub_results <- as.numeric(df$CHR) == chrom & as.numeric(df$BP) >= BPstart & as.numeric(df$BP) <= BPend;
     print(head(sub_results))
     region_results <- subset(results, sub_results)
     return(region_results)
    }

Here are the prints to the console containing the region information being used:

[1] 1
[1] 113.308
[1] 115.158

Here is the print from the subset (region_results):

     GENE CHR     SNP    EMP1  NP        BP          SNP_IM  SNP_LZ
3238 AP3S1   5    rs26538 1.00000   6 115178395         rs26538    rs26538
3239 AP4B1   1  rs1217401 1.00000  46 114438951 imm_1_114240474  rs1217401
3240 AP4B1   1  rs1217402 1.00000  41 114440258 imm_1_114241781  rs1217402
3241 AP4B1   1  rs3789613 1.00000 297 114443035 imm_1_114244558  rs3789613
3242 AP4B1   1  rs7523862 1.00000 297 114443419 imm_1_114244942  rs7523862
3243 AP4B1   1 rs17464525 1.00000 148 114443899 imm_1_114245422 rs17464525

As you can see, there is a row in the subset containing a marker in chromosome 5. What am I doing wrong? Thank you in advance. edit: Here is the call to the function with the stuff before it:

     write.genelist <- function(table_loc, region.info, out_folder,yank_loc){
      region.ID = as.character(region.info[[1]])
      out_name = paste0(region.ID,"_genes.list")
      region_folder = file.path(out_folder, region.ID)
      out_loc <- file.path(region_folder,out_name, fsep = .Platform$file.sep)
      results <- read.table(table_loc, T,strip.white = TRUE)
      gene_region_results <- subset.by.region(results,region.info)
      ...
     }

Upvotes: 1

Views: 706

Answers (1)

Gavin Simpson
Gavin Simpson

Reputation: 174788

I would use [ to subset in a function not subset(). See ?subset for why.

subset.by.region<- function(results, df, region.info, expansion=0){
    MBstart = as.numeric(region.info[[3]]) - expansion
    MBend = as.numeric(region.info[[4]]) + expansion
    chrom =as.numeric(region.info[[2]])
    print(chrom)
    print(MBstart)
    print(MBend)
    BPstart <- MBstart  * 1e6
    BPend <- MBend  * 1e6
    sub_results <- as.numeric(df$CHR) == chrom & 
        as.numeric(df$BP) >= BPstart & as.numeric(df$BP) <= BPend
    print(head(sub_results))
    results[sub_results, ]
}

I would also pass in results not rely on it being found in the global environment.

Upvotes: 1

Related Questions