No LD information

[Paul Simon, Mr]

Good morning,

My name is Paul Simon and I am a doctoral student at McGill university.

I am currently trying to use locuszoom to generate LD plots for specific SNPs that I plan on presenting in a publication.

I have a total of 28 SNPs.

Out of these I was able to obtained 24 plots with no issues.

I ran into some problems for 4 SNPs:

- rs66801926

- rs34582534

- rs5810734

- rs150233516

The error message I received for these is the following one:
"Genotypes from VCF have more than 4 possible states".

Our genotype file clearly shows that we have the appropriate number of states for those SNPs.

I was wondering if there was a way to solve this problem?

Cordially,

Paul Simon, BSc, MSc, PhD Candidate

Co-President - Junior Enterprise Desautels

1-(514)-996-6663

paul....@mail.mcgill.ca

[Matthew Flickinger]

What command(s) exactly did you run to attempt to produce the plots? Were you using the online tool or the stand alone program? If you are able to make a minimal, reproducible example, that would make it much easier to help you.

Matthew Flickinger, Ph.D.
Senior Applications Programmer
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+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Paul SIMON]

Hello Matthew,

Thank you for answering my question.

I attempted to plot my data with the online tool and the through a bash script.

When using the bash script, the source file is 1000G_March 2012 with the build h19. The plot is produced but is grey and with the message "No usable LD information for reference SNP".

When using the online tool, I upload my GWAS data (.assoc file) and receive the same output. Locuszoom is able to generate the plot but indicates at the bottom "No usable LD information for reference SNP".

I'm joining the screen capture of when I run it on the online tool.

I could also send you the log file from when I run it on bash.

Thanks for your help

Paul

To unsubscribe from this group and stop receiving emails from it, send an email to locuszoom+...@googlegroups.com.

[Matthew Flickinger]

Hello Paul,

I took a look at the SNPs you listed. All of those appear to be deletions. All the files that locuszoom uses for plotting contain only bi-allelic single base pair SNPs. We've filtered out all deletions.This is the reason that no usable LD information is found.

If you need to make colored plots with these SNPs, you can download the 1000G VCFs for the region of interest (you can use tabix for this: eg "tabix ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/20130502/ALL.chr6.phase3_shapeit2_mvncall_integrated_v5a.20130502.genotypes.vcf.gz 6:157160000-157165000 > grab.20130502.vcf") and make sure your SNPs of interest are in there. Then you can pass a custom "--ld-vcf" parameter to the stand alone locuszoom application (see http://genome.sph.umich.edu/wiki/LocusZoom_Standalone#Supply_VCF_files_for_calculating_LD). This cannot currently be done with the website.

Hope this helps.

-Matthew

To unsubscribe from this group and stop receiving emails from it, send an email to locuszoom+unsubscribe@googlegroups.com.

[PaulHG]

Good morning Matthew,

Thank you very much for your help.

I'll give this a try today.

Paul Simon

You received this message because you are subscribed to a topic in the Google Groups "locuszoom" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/locuszoom/Yam0lKY2Xag/unsubscribe.
To unsubscribe from this group and all its topics, send an email to locuszoom+...@googlegroups.com.

[alexi...@gmail.com]

Hi Matthew,

May I ask whether using 1000G VCFs can avoid indel SNPs being colored as grey? I have used 1000G phase 3 panel for LD reference (--ld-vcf) and it was the same panel I use for imputation, but I still got grey dots on the plots. Is the only explaination that the grey dot SNPs are indel so locuszoom can't handle them? Or is there any other possible reason?

Many thanks.

Alexis

在 2017年1月31日星期二 UTC+11上午9:54:28，Matthew Flickinger写道：

[Matthew Flickinger]

Hello Alexis,

It's hard to say what's going on with out a clear example. There should be no reason INDELs should always show up as grey. Here's an example that works for me (this are unix commands that assume R and tabix are installed as well as locuszoom)

tabix -h ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/20130502/ALL.chr6.phase3_shapeit2_mvncall_integrated_v5a.20130502.genotypes.vcf.gz 6:157160000-157165000 > test_input.vcf

cat <<EOT > make_results.R

set.seed(100)

dd <- read.table("test_input.vcf", stringsAsFactors=FALSE)

results <- data.frame(CHROM = dd\$V1,

    BEGIN = dd\$V2, END = dd\$V2,

    MARKER_ID = dd\$V3, NS=NA,

    AC=NA, CALLRATE=NA, MAF=NA,

    PVALUE = runif(nrow(dd)), SCORE=NA)

names(results)[1] <- "#CHROM"

write.table(results, "test_results.txt", quote=F, sep="\t", row.names=F)

EOT

Rscript make_results.R

bgzip test_input.vcf

tabix test_input.vcf.gz

locuszoom --epacts test_results.txt --ld-vcf test_input.vcf.gz --chr 6 --start 157160000 --end 157165000

It appears to me that the INDELS in the the input are being properly colored (though this is a very boring plot). 

Matthew Flickinger, Ph.D.
Senior Applications Programmer
Center for Statistical Genetics
University of Michigan

To unsubscribe from this group and stop receiving emails from it, send an email to locuszoom+unsubscribe@googlegroups.com.

[Aaron Aguhob]

Hi Paul,

I was wondering if you were able to fix your issue? I am running into the same error "No usable LD information for reference SNP" when I try to supply my own vcf file for use. Since my data is of mixed population, I wanted to use my own vcf to calculate LD rather than using the references provided by locuszoom. It appears to be just an issue with my vcf since it does produce plots (except the plots are grey).

Thank you,

Aaron Aguhob

[Wander Purple]

Same error with my own vcf file, reporting "index SNP position *** does not exist in file.." .

I check the SNP position and its flanking region with tabix. They all exists. Seems code of "locuszoom/src/vcf_ld.py" at 85 line cannot process properly

Is there any remedy?

Thanks

Wander Purple

在 2018年2月16日星期五 UTC+8上午6:10:29，Aaron Aguhob写道：

[DamienTan]

I got the same error in my vcf file:  "index SNP position *** does not exist in file..".

I noticed that there were more than 2 SNPs in the same position 

just like that：

7       138084256       chr7:138084256_C/CTT
7       138084256       chr7:138084256_C/CTTT
7       138084256       chr7:138084256_CT/C

Maybe for these SNPs which include multi records in the same position, it will cause all the dots grey.

Is that any advice that can help me solve this problem?

Thanks 

Damien Tan