hg38 mappability files for BiC-seq2

[Bogdan Tanasa]

Dear all, 

thank you for you work and for making the mappability files available for hg38. I would like to use hg38 mappability files for BIC-seq2 : http://compbio.med.harvard.edu/BIC-seq/

please could you advise on how I shall change the 50mer or 100mer MAPPABILITY FILES in order to make the FORMAT compatible  with BiC-seq2, in the sense that : 

1 --  BIC-seq2 mappability file that is needed (per chromosome) lists NON-OVERLAPPING REGIONS : 

an example from CHR1 : 

"10027 10032

10052 10062

10065 10534

10707 10712

10723 10814

15538 15541

15809 15902"

2 -- on UMAP website, the file k100 (here for hg19) lists OVERLAPING REGIONS : 

an example from CHR1 :

chr1 10009 10721 k100 1 +

chr1 10701 10812 k100 1 +

chr1 10717 10913 k100 1 +

chr1 12979 13170 k100 1 +

chr1 13227 13425 k100 1 +

chr1 13408 13571 k100 1 +

thanks a lot ! appreciate your help !

-- bogdan

[Mehran Karimzadeh]

Hello Bogdan,

As you noted, the single-read mappability files have overlapping regions.

There is no difference between an overlapping and non-overlapping region in terms of single-read mappability.

One suggestion is to use a tool like bedtools merge to unify these regions.

A better solution is to filter the bedGraph file of multi-read mappability regions by a particular threshold and generate a BED file from there.

The reason this is better is that the single-read mappability BED files include all regions which are uniquely mappable, regardless of the multi-read mappability score.

Hope this helps

Mehran

--
You received this message because you are subscribed to the Google Groups "Ubismap" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ubismap+u...@googlegroups.com.
To post to this group, send email to ubi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ubismap/87047b9c-70e6-4b7c-81ed-4e39dc3a987a%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Bogdan Tanasa]

Hi Mehran, 

thank you, it is very very helpful. we wish you the best, 

-- bogdan