search

bulik / ldsc

LD Score Regression (LDSC)
GNU General Public License v3.0
628 stars 339 forks source link

Issues with ldscores from 2 LDSC_Seg annotations #83

Open naglem opened 7 years ago

naglem commented 7 years ago

Hi,

I've successfully replicated results using some of the ldscores from the most recent biorxiv paper for partitioned heritability (Heritability enrichment of specifically expressed genes identifies disease-relevant tissues and cell types) including the gtex and cahoy annotations. However, I run into the same error for the Franke and ImmGen data: line 286, in cell_type_specific for (name, ct_ld_chr) in [x.split() for x in open(args.ref_ld_chr_cts).readlines()]: ValueError: need more than 0 values to unpack Any help on this would be great, thanks.

rkwalters commented 7 years ago

Hi Michael, The error message suggests an issue with reading Franke.ldcts (or ImmGen.ldcts), specifically that it’s hitting a line that doesn’t have the expected two fields.

I haven’t been able to replicate the error yet. The line reporting the error parses Franke.ldcts without any issues for me. I’m in the process of a full test run with the Franke data, but it has made it past the point where this error should arise and is still running smoothly.

Have you done anything to modify the .ldcts files? On the small chance Franke.ldcts somehow got corrupted, could you try redownloading that file and rerunning?

Cheers, Raymond

On May 15, 2017, at 4:20 PM, Michael Nagle notifications@github.com wrote:

Hi,

I've successfully replicated results using some of the ldscores from the most recent biorxiv paper for partitioned heritability (Heritability enrichment of specifically expressed genes identifies disease-relevant tissues and cell types) including the gtex and cahoy annotations. However, I run into the same error for the Franke and ImmGen data: line 286, in cell_type_specific for (name, ct_ld_chr) in [x.split() for x in open(args.ref_ld_chr_cts).readlines()]: ValueError: need more than 0 values to unpack Any help on this would be great, thanks.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/bulik/ldsc/issues/83, or mute the thread https://github.com/notifications/unsubscribe-auth/AILEvc7aOFrG26RWfEsOMkTGmK26l6Ojks5r6LOjgaJpZM4NbpAE.

naglem commented 7 years ago

Thanks, I did modify the .ldcts files to reflect my paths. I went back and reformatted them again, and works like a charm.

Thanks for the assist.