LD not colored in LocalZoom

[Jaehyun Park]

Hello,

 

I would like to make an inquiry about problems in generating plots with LocalZoom.

 

I uploaded GWAS summary statistics files with the chromosome/position/ref & alt allele information for each variant.

The plot was successfully generated, but no LD information was shown in a color scale (all other variants were colored in ‘grey’).

Even though the variants were also detected with GWAS Catalog hits, the LD information corresponding to those variants was not shown.

 

Below are the examples of the variants that did not show LD information (GRCh38-based):

1:109275684

4:147480038

 

When I searched for the problem online, I found some comments saying that the 1000G reference panel is a bit old and some variants may be missing.

I think it is possible also for our case, but since the comment was written a several years ago, I am not sure if this is a still valid explanation.

 

I would like to ask the possible reason of the problem and whether there is something I can do to solve it.

 

I would appreciate it if you could give me a response.

 

Thank you.

 

 

Best regards,

Jaehyun Park.

[Andy Boughton]

The following might be helpful. LocalZoom and my.locuszoom.org share a common codebase, so the advice applies to both tools:

https://my.locuszoom.org/about/#missing-ld

For example, I notice that your variants don't specify ref or alt alleles (aligned with the human reference genome). This is required for calculating LD.

(other matches, like gwas catalog, may be more lenient)

-Andy Boughton
abo...@umich.edu

Applications Programmer/Analyst, Lead
Center for Statistical Genetics
University of Michigan

--
You received this message because you are subscribed to the Google Groups "locuszoom" group.
To unsubscribe from this group and stop receiving emails from it, send an email to locuszoom+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/locuszoom/8ecbaf02-d054-4725-933a-11d699017cacn%40googlegroups.com.

[Jaehyun Park]

Thank you for the response.

I checked the link and tried specifying ref and alt alleles.

However, I still cannot see LD information in my plot.

For example, I specified the reference allele of 1:109275684 as G and the alternative allele as T (as in GRCh38),

but the plot did not show LD information.

I also tried specifying ref as T and alt as G, but it did not show any change.

I think it may be possible that this variable is not in the LD reference panel.

If there is another possible reason of the problem, I would appreciate it if you could let me know.

Thank you.

Best regards,

Jaehyun Park.

2024년 3월 18일 월요일 오후 1시 43분 25초 UTC-5에 Andy Boughton님이 작성:

[Andy Boughton]

The known reasons are all spelled out in the FAQ; might be worth checking the 1000G panel to see if it's a known variant.

(incidentally, the order of "T" vs "G" isn't a matter of guess work: these are the ref and alt alleles exactly as spelled out in the genome build)

Interestingly, a test query to the LD server says there is known LD information relative to the variant. You might see other variants fail to match if most of the variants in your plot are very rare (a mix of grey and colored points).

curl "https://portaldev.sph.umich.edu/ld/genome_builds/GRCh38/references/1000G/populations/ALL/variants?correlation=rsquare&variant=1:109275684_G/T&chrom=1&start=108775684&stop=109775684"

-Andy Boughton
abo...@umich.edu

Applications Programmer/Analyst, Lead
Center for Statistical Genetics
University of Michigan

To view this discussion on the web visit https://groups.google.com/d/msgid/locuszoom/950f841a-0f28-4a69-ad73-aee2cd52e91fn%40googlegroups.com.

[Jaehyun Park]

I would like to report that the LD was successfully colored when I used LocusZoom, not LocalZoom.

I apologize for bothering with the multiple inquiries.

Thank you.

Best regards,

Jaehyun Park.

2024년 3월 18일 월요일 오후 2시 34분 8초 UTC-5에 Andy Boughton님이 작성: