search

DiltheyLab / HLA-LA

Fast HLA type inference from whole-genome data
GNU General Public License v3.0
120 stars 40 forks source link

DRB4*03:01N #84

Closed afadda1 closed 11 months ago

afadda1 commented 1 year ago

Hi, I used HLA-LA to type hundreds of WGS and noticed that i'm getting a higher frequency for the null allele DRB4*03:01N than reported before. This allele has a big deletion that includes exons 2 & 3, the tow exons HLA-LA relies on for typing. I used HLA-HD to check my results, and it reports a different non-null allele. I'm wondering if you have a word of caution regarding this particular allele, or should i trust HLA-LA results? thanks

AlexanderDilthey commented 1 year ago

I think I would suspect a systematic error. Does the proportionkMersCovered metric look good for these calls?

afadda1 commented 1 year ago

It’s -1 but not sure what it means I’m attaching the bestguess file

From: Alexander Dilthey @.> Reply-To: DiltheyLab/HLA-LA @.> Date: Wednesday, 11 January 2023 at 15:47 To: DiltheyLab/HLA-LA @.> Cc: afadda1 @.>, Author @.**> Subject: Re: [DiltheyLab/HLA-LA] DRB403:01N (Issue #84)

I think I would suspect a systematic error. Does the proportionkMersCovered metric look good for these calls?

— Reply to this email directly, view it on GitHubhttps://github.com/DiltheyLab/HLA-LA/issues/84#issuecomment-1378860773, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AUE7GLPJUYFMG4IMTWWLWCDWR3BW5ANCNFSM6AAAAAATXFYNKU. You are receiving this because you authored the thread.Message ID: @.***>

Locus Chromosome Allele Q1 Q2 AverageCoverage CoverageFirstDecile MinimumCoverage proportionkMersCovered LocusAvgColumnError NColumns_UnaccountedAllele_fGT0.2 perfectG A 1 A03:02:01G 1 -4 61.5299 55 50 1 4.97772e-05 0 1 A 2 A32:01:01G 1 -4 61.5299 55 50 1 4.97772e-05 0 1 B 1 B35:01:01G 1 -5 72.2549 63 57 1 0 0 1 B 2 B44:02:01G 1 -5 72.2549 63 57 1 0 0 1 C 1 C04:01:01G 1 -2 59.598 55 51 1 0 0 1 C 2 C05:01:01G 1 -2 59.598 55 51 1 0 0 1 DQA1 1 DQA101:01:01G 1 -2 28.6586 24 21 1 0.000280269 0 1 DQA1 2 DQA105:01:01G 1 -2 28.6586 24 21 1 0.000280269 0 1 DQB1 1 DQB103:01:01G 1 -3 84.7741 75 67 1 0 0 1 DQB1 2 DQB105:01:01G 1 -3 84.7741 75 67 1 0 0 1 DRB1 1 DRB110:01:01 1 -2 47.5091 39 36 1 7.65404e-05 0 0 DRB1 2 DRB111:01:01G 1 -2 47.5091 39 36 1 7.65404e-05 0 1 DPA1 1 DPA101:03:01G 1 -0 75.2683 66 63 1 0 0 1 DPA1 2 DPA101:03:01G 1 -0 75.2683 66 63 1 0 0 1 DPB1 1 DPB102:01:02G 1 -1 71.1932 60 58 1 0 0 1 DPB1 2 DPB104:01:01G 1 -1 71.1932 60 58 1 0 0 1 DRB3 1 DRB302:02:01G 1 -11 38.1047 33 32 1 0.589389 0 1 DRB3 2 DRB302:02:01G 0.998297 -11 38.1047 33 32 1 0.589389 0 1 DRB4 1 DRB403:01N 1 -0 58 58 58 -1 0 0 1 DRB4 2 DRB403:01N 1 -0 58 58 58 -1 0 0 1 E 1 E01:01:01G 1 -2 54.1996 42 33 1 6.75836e-05 0 1 E 2 E01:03:02G 1 -2 54.1996 42 33 1 6.75836e-05 0 1 F 1 F01:01:01G 1 -0 61.8425 51 46 1 0 0 1 F 2 F01:01:01G 1 -0 61.8425 51 46 1 0 0 1 G 1 G01:01:01G 1 -0 71.5293 59 50 1 0 0 1 G 2 G01:01:12 1 -0 71.5293 59 50 1 0 0 1

afadda1 commented 1 year ago

Looking forward to your comments

From: Alexander Dilthey @.> Reply-To: DiltheyLab/HLA-LA @.> Date: Wednesday, 11 January 2023 at 15:47 To: DiltheyLab/HLA-LA @.> Cc: afadda1 @.>, Author @.**> Subject: Re: [DiltheyLab/HLA-LA] DRB403:01N (Issue #84)

I think I would suspect a systematic error. Does the proportionkMersCovered metric look good for these calls?

— Reply to this email directly, view it on GitHubhttps://github.com/DiltheyLab/HLA-LA/issues/84#issuecomment-1378860773, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AUE7GLPJUYFMG4IMTWWLWCDWR3BW5ANCNFSM6AAAAAATXFYNKU. You are receiving this because you authored the thread.Message ID: @.***>

AlexanderDilthey commented 11 months ago

Hi @afadda1,

You are right, the metric proportionkMersCovered is affected by the deletion. I would recommend caution when intepreting any results associated with DRB4*03:01N.

Best wishes

Alex