Hello everyone

I hope you are safe and good

I have genomic ranges below in which minor allele = 0 means loss of Heterozygosity. (LOH)

Chr start         end
1   10583       863511
1   12841835    12854479
1   17231299    17232877
1   120380739   142540462
1   142540484   142684781
1   142685823   143462598
1   143466928   143533092
1   143534046   144064964
1   145079160   145110146
1   145121086   145400508

I have another data frame like below where I have position of mutations in each genes

> head(mut[,c(2:7)])
  Chromosome Start_Position End_Position Reference_Allele Tumor_Seq_Allele2
1       chr1       89616151     89616151                -                 T
2       chr6       51909815     51909815                A                 -
3      chr16       20556547     20556547                -                 G
4      chr17       66992092     66992092                -                 A
5      chr20       13251339     13251339                -                 T
6      chr20       35929770     35929771               TC                 -
  Hugo_Symbol
1        GBP7
2       PKHD1
3      ACSM2B
4       ABCA9
5        ISM1
6      MANBAL
>

I want to know which genes also have los of Heterozygosity based on the first data frame

Thank you for any help

Hello again Fereshteh,

This can be achieved via findOverlaps(), which is part of the GenomicRanges package. Here is a reproducible example:

1, load package

require(GenomicRanges)

2, create gr1 object

gr1 <- makeGRangesFromDataFrame(
    data.frame(
        chr = rep('chr1',3),
        start = c(1,100,500),
        end = c(50,150,1500),
        feature = c('feat1', 'feat2', 'feat3')),
        keep.extra.columns = TRUE)

gr1

GRanges object with 3 ranges and 1 metadata column:
      seqnames    ranges strand |     feature
         <Rle> <IRanges>  <Rle> | <character>
  [1]     chr1      1-50      * |       feat1
  [2]     chr1   100-150      * |       feat2
  [3]     chr1  500-1500      * |       feat3
  -------
  seqinfo: 1 sequence from an unspecified genome; no seqlengths

3, create gr2 object

gr2 <- makeGRangesFromDataFrame(
    data.frame(
        chr = rep('chr1',3),
        start = c(100,500,1001),
        end = c(200,1000,1050),
        gene = c('ENSG2', 'ENSG3', 'ENSG4')),
        keep.extra.columns = TRUE)

gr2

GRanges object with 3 ranges and 1 metadata column:
      seqnames    ranges strand |        gene
         <Rle> <IRanges>  <Rle> | <character>
  [1]     chr1   100-200      * |       ENSG2
  [2]     chr1  500-1000      * |       ENSG3
  [3]     chr1 1001-1050      * |       ENSG4
  -------
  seqinfo: 1 sequence from an unspecified genome; no seqlengths

4(a), find any features in gr1 that overlap any in gr2

gr1[queryHits(findOverlaps(gr1, gr2, type="any")),] 

GRanges object with 3 ranges and 1 metadata column:
      seqnames    ranges strand |     feature
         <Rle> <IRanges>  <Rle> | <character>
  [1]     chr1   100-150      * |       feat2
  [2]     chr1  500-1500      * |       feat3
  [3]     chr1  500-1500      * |       feat3
  -------
  seqinfo: 1 sequence from an unspecified genome; no seqlengths

4(b), same as 4(a), but add on [to the output] those features in gr2 that are overlapped

data.frame(
  gr1[queryHits(findOverlaps(gr1, gr2, type="any")),],
  gr2[queryHits(findOverlaps(gr2, gr1, type="any")),])

  seqnames start  end width strand feature seqnames.1 start.1 end.1 width.1
1     chr1   100  150    51      *   feat2       chr1     100   200     101
2     chr1   500 1500  1001      *   feat3       chr1     500  1000     501
3     chr1   500 1500  1001      *   feat3       chr1    1001  1050      50
  strand.1  gene
1        * ENSG2
2        * ENSG3
3        * ENSG4

This should make this task very easy for you.

By the way, in your mut, object, I would change 'Chromosome', 'Start_Position', and 'End_Position' to 'chromosome', 'start' and 'end'; otherwise, GenomicRanges will not automatically recognise these columns.

Kevin