How to get promoter coordinates of hg19 from UCSC genome browser ?

Question

[jack]

Hi all,

I need to get Promoter coordinates of all genes in human genome from hg19 assembly.

Is it possible to get it from UCSC table? I tried, but I was not successful.

Would someone can help me with that?

Answer 1

[Giovanni M Dall'Olio]

This should be a simple question, but in reality there are many approaches because there are multiple definitions of promoters.

The simplest way to do it is to go the the HG19 folder of the UCSC FTP site and download the upstream1000.fa.gz file, containing the sequence of the promoters for all the human genes.

If you are familiar with R you can do it using the brand new AnnotationHub interface from BioConductor. For more information, follow the tutorial here. In particular this code is based on this video.

> source("http://bioconductor.org/biocLite.R")
> biocLite("GenomicRanges")
> biocLite("AnnotationHub")
> biocLite("rtracklayer")
> library("GenomicRanges")
> library("AnnotationHub")
>
> qhs = query(ahub, c("RefSeq", "Homo sapiens", "hg19"))
> genes = qhs[[1]]
> proms = promoters(genes)

UCSC track 'refGene'
UCSCData object with 50066 ranges and 5 metadata columns:
                       seqnames               ranges strand   |         name     score     itemRgb                thick
                          <Rle>            <IRanges>  <Rle>   |  <character> <numeric> <character>            <IRanges>
      [1]                  chr1 [66997825, 67000024]      +   |    NM_032291         0        <NA> [67000042, 67208778]
      [2]                  chr1 [ 8376145,  8378344]      +   | NM_001080397         0        <NA> [ 8378169,  8404073]
      [3]                  chr1 [50489427, 50491626]      -   |    NM_032785         0        <NA> [48999845, 50489468]
      [4]                  chr1 [16765167, 16767366]      +   | NM_001145277         0        <NA> [16767257, 16785491]
      [5]                  chr1 [16765167, 16767366]      +   | NM_001145278         0        <NA> [16767257, 16785385]
      ...                   ...                  ...    ... ...          ...       ...         ...                  ...
  [50062] chr19_gl000209_random     [ 55209,  57408]      +   |    NM_002255         0        <NA>     [ 57249,  67717]
  [50063] chr19_gl000209_random     [ 44646,  46845]      +   | NM_001258383         0        <NA>     [ 57132,  67717]
  [50064] chr19_gl000209_random     [ 96135,  98334]      +   |    NM_012313         0        <NA>     [ 98146, 112480]
  [50065] chr19_gl000209_random     [ 68071,  70270]      +   | NM_001083539         0        <NA>     [ 70108,  83979]
  [50066] chr19_gl000209_random     [129433, 131632]      +   |    NM_012312         0        <NA>     [131468, 145120]
                                                    blocks
                                             <IRangesList>
      [1] [    1,   227] [91706, 91769] [98929, 98953] ...
      [2]       [   1,  102] [6222, 6642] [7214, 7306] ...
      [3]       [   1, 1439] [2036, 2062] [6788, 6884] ...
      [4]       [   1,  182] [2961, 3061] [7199, 7303] ...
      [5]       [   1,  104] [2961, 3061] [7199, 7303] ...
      ...                                              ...
  [50062]       [   1,   80] [ 280,  315] [1182, 1466] ...
  [50063] [    1,    86] [10414, 10643] [10843, 10878] ...
  [50064]       [   1,   46] [1523, 1557] [4002, 4301] ...
  [50065]       [   1,   71] [1071, 1106] [1851, 2135] ...
  [50066]       [   1,   69] [ 862,  897] [3334, 3633] ...

Answer 2

[Chirag Nepal]

You should be able to download promoter table from UCSC browser. Alternative you can download the gene coordinates in .bed format. Define upstream and downstream region in your assigned promoter region, which is generally 500 or 1000 nucleotides.

upsteam=500

downstream=500

cat ucscRefseq.bed | awk '{ if ($6 == "+") { print $1,$2-'$upstream', $2+'$downstream', $4, $5, $6,$7,$8,$9,$10,$11,$12 } else if ($6 == "-") { print $1, $3-'$upstream', $3+'$downstream', $4,$5,$6,$7,$8,$9,$10,$11,$12 }}' > promoter.bed

[rbronste]

Trying the awk script and getting the following error:

awk: cmd. line:1:
^ unexpected newline or end of string

Any ideas? Thanks!

[Chirag Nepal]

There was one closing bracket missing, edited it, try it now.