Beagle 5.2 Imputation Issue, understanding output

Question

[June]

Hey everyone,

I'm having some trouble with the imputation step in my GBS pipeline. I'm following the FastGBS pipeline basically as written up until this point.

Before running Beagle, i've performed a filtering step using vcftools (though I get the same error whether or not I do this; I just figure it's possible it's an earlier step that something has gone wrong) -

vcftools --vcf FastGBS_platypus_Phapa.vcf --remove-filtered-all --max-missing 0.2 --remove-indels --mac 1 --min-alleles 1 --max-alleles 2 --recode --out FastGBS_p$

With this command I don't get an error but do get some warnings I'm not sure what to do with

Warning: Expected at least 2 parts in INFO entry: ID=Source,Number=.,Type=String,Description="Was this variant suggested by Playtypus, Assembler, or from a VCF?">
Warning: Expected at least 2 parts in INFO entry: ID=Source,Number=.,Type=String,Description="Was this variant suggested by Playtypus, Assembler, or from a VCF?">
Warning: Expected at least 2 parts in FORMAT entry: ID=GL,Number=.,Type=Float,Description="Genotype log10-likelihoods for AA,AB and BB genotypes, where A = ref and B = variant. Only appl$

Following this step I run beagle using the following command

java -Xmx25000m -jar /home/jec19078/beagle.21Apr21.304.jar gt="${outplat}".recode.vcf out="${outplat}"_recode_imputed

During the beagle run, it seems to run fine for the first few scaffolds and then it will get hung up on a scaffold with only a single study marker.

Window 9 [scaffold_1010:131483-131483]    
Reference markers:                    1  
Study     markers:                    1   
Exception in thread "main" java.lang.IllegalArgumentException: 131483  
        at vcf.MarkerMap.meanSingleBaseGenDist(MarkerMap.java:97)  
        at phase.FixedPhaseData.markerMap(FixedPhaseData.java:169)   
        at phase.FixedPhaseData.<init>(FixedPhaseData.java:116)   
        at main.Main.phaseStage1Variants(Main.java:187)   
        at main.Main.phaseTarg(Main.java:181)   
        at main.Main.phaseAndImpute(Main.java:171)   
        at main.Main.main(Main.java:126)   
        !!! There is a problem at the imputation step

I hope this is enough information to get some help, I'm happy to provide more info as to how I did certain things so I can move forward.

Answer 1

[Matt]

Did you resolve this? I'm getting the same error on a scaffold that reports blank (instead of 0) reference and study markers.

[Matt]

The problem is with singleton sites (note the number of markers in your window). The solution is to exclude these scaffolds using the excludemarkers=file option. For example, if you have 5 scaffolds that each have one SNP, you'd list those sites in a file:

scaffold00045:314
scaffold00049:4600
scaffold00068:7534
scaffold00091:1165
scaffold00110:4799

Thanks to the software author for helping me resolve the issue.