This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Regions of extreme GC content

Hello for CNV analysis QC, I am looking for a reliable bed file of regions with extreme (>90%, <10%) GC content.

Any idea were I can find such file? I tried UCSC but it gives it in bins of 5bp, which is not very convenient.

Thanks

gc-content

1 answer

It's not too difficult to generate a bed file of GC content along the genome. You just need the reference fasta file and a genome file giving the length of the chromosomes. Then with bedtools, first create sliding windows along the genome and for each window calculate the %GC, then use e.g. awk to get rows where %CG is above/below a threshold, something like:

bedtools makewindows -g hg19.genome -w 1000 \
| nucBed -fi hg19.genome.fasta -bed \
| awk '$5 > 0.9 || $5 < 0.1'

See 'nucBed -h' for the output format, I think %GC is going to be in column 5, not sure though.

Did you notice the | in the command? Output of bedtools makewindows is directly being piped into nucBed (it is in the bed format).

Actually I have an error:

"Less than the req'd two fields were encountered in the genome file (hg19.genome) at line 1."

and my hg19.genome is:

chr1    249250621
chr2    243199373
chr3    198022430
chr4    191154276
chr5    180915260
chr6    171115067
chr7    159138663
chrX    155270560
chr8    146364022
chr9    141213431
.
.
.

What turned out to be the problem?

I changed "chr1" to "1" and .... . then my error was fixed.

Do you know how can I compute mappability? I want to fix mappability bias and I need mappability like gc content.

Log in to answer this question.