CCR BED files in hg38 genome build

[enriquea]

Hi there,

Great work!

I was wondering if you have the CCRs in the new genome build coordinates (hg38).

Thanks in advance.

[jimhavrilla]

Hey, Sorry, not as of yet. I have not gotten around to re-running the CCR model on the new build. It's sadly pretty low priority for me at the moment. Some time in the future I plan to do so, probably around July. I will keep you posted, and keep this issue open.

…

On Wed, Mar 4, 2020 at 1:36 PM Enrique Audain ***@***.***> wrote: Hi there, Great work! I was wondering if you have the CCRs in the new genome build coordinates (hg38). Thanks in advance. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#4?email_source=notifications&email_token=ABSDYBDD6NQZLEJEBEYQO53RF2NT5A5CNFSM4LBQFW4KYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4ISP2WHA>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABSDYBH6LR4C7J7WMCVY7WTRF2NT5ANCNFSM4LBQFW4A> .

[enriquea]

Hi Jim,

Thanks for your quick reply. In the meantime, do you think it is a valid approach to lift them over to the new genome assembly?

I would expect comparable results, but I'm not sure.

Best,

Enrique

[jimhavrilla]

Personally, I've heard mixed things about lifting over from people who have done it with VCF data. In regions with poor mappability you may see as low as 60% retention, and in a well-covered region of the genome, 85+%. So if you're really in need of it, it could work in a pinch just take the coordinates with a strong grain of salt. I can create the new regions without too much trouble, it's just that I have to do quality control of them before making an official release which will take some time. Either way good luck with whatever you need it for or decide to do, and the moment they're updated I will say so in this issue. Jim Havrilla

…