search

DCGenomics / DangerTrack

MIT License
13 stars 3 forks source link

Scores #5

Open berntpopp opened 7 years ago

berntpopp commented 7 years ago

Hi,

apparently the scores got changed in the .bed files without documentation:

zcat /illumina/bernt/tools/DangerTrack/dangertrack.bed.gz | more

chr1    0       5000    chr1:0-5000     500
chr1    10000   15000   chr1:10000-15000       458
chr1    100000  105000  chr1:100000-105000      434
chr1    1000000 1005000 chr1:1000000-1005000    211
chr1    10000000        10005000        chr1:10000000-10005000  73
chr1    100000000       100005000       chr1:100000000-100005000        17
chr1    100005000       100010000       chr1:100005000-100010000        24
chr1    100010000       100015000       chr1:100010000-100015000        10

The cutoff now seems to be >=500 as 48925/619151 ~ 7.9% ("not trustworthy" region in paper). Can you confirm please?

Best regards, Bernt

bballew commented 6 years ago

I'm also interested in this issue. The DangerTrack publication describes scores between 0 and 1; the provided bed file does not match this description. Can you please explain?

Thank you! Bari

fritzsedlazeck commented 6 years ago

sorry for the late reply. It is true the current code scales it differently. I will check with Igor and try to report back. Cheers Fritz

fritzsedlazeck commented 6 years ago

@igordot

igordot commented 6 years ago

We scaled the scores between 0 and 1 for the publication. The output BED file score field is scaled 0-1000 to better comply with the BED file specifications. The only difference is the shifting of the decimal point.