Running Beagle for imputation

Question

[nour.hadjz]

Hello, i work on a Macbook Pro with 16Go and i want to run Beagle using this syntax:

java -Xmx16g -jar beagle.27Jan18.7e1.jar get=sortieMSL.recode.vcf  out= result   ref=chr22.1kg.vcf.gz  chrom=22  impute=true

but it doesn't work i have an Error message :

unrecognized parameter

i know that i have something wrong but i don't know how to do!

can you please help me with my syntax ?!

[WouterDeCoster]

out= result

Just a guess, what if you remove the space here?

[nour.hadjz]

I removed the space but it does not work

Answer 1

[Kevin Blighe]

Did you check the manual: Beagle 4.1 ?

Try this:

java -Xmx16g -jar beagle.27Jan18.7e1.jar gt=sortieMSL.recode.vcf out=result ref=chr22.1kg.vcf.gz chrom=22 impute=true

Other things about which to be careful:

• if Beagle does not like compressed VCFs, then you will have to decompress your chr22.1kg.vcf.gz file if Beagle accepts compressed or uncompressed VCFs, then your compressed VCFs should be compressed with bgzip and then tab-indexed with tabix -p vcf Variant.vcf.gz
• for the chrom parameter, be sure that your chromosomes in your VCF input files are labelled in the same way, i.e., '22' and not 'chr22'

PS - you should not require 16GB RAM if you are just looking at chr22.

Kevin

[nour.hadjz]

yes i saw the manual . what size of memory and required for chr 22 thank you

[Kevin Blighe]

2 or 4GB?

[nour.hadjz]

Hello, it's me again. after trying to impute with Beagle i have a message : Missing line (#CHROM ...) after meta-information lines.

do you have any suggestion please ?!

[WouterDeCoster]

You should check if the header of your vcf is intact. That header indeed starts with #CHROM

[nour.hadjz]

in my VCF file i have the header #CHROM actually !

[Kevin Blighe]

Are you running this on Windows?

[nour.hadjz]

no i m not running it on windows , im working on a Mac! what could be the problem??

[nour.hadjz]

i think i have this error message because the input VCF file does not conform to the VCF format specification so do you know un validator programs that i can use to resolve my problem please ?!

thank you

[Kevin Blighe]

The process of zipping your VCF with bgzip [VCF] and then tab-indexing with tabix -p vcf [VCF.GZ] can sometimes fix errors with formatting. Alternatively, you can just paste the VCF header and some variants here, so that we can take a look at it.

[nour.hadjz]

Hello, Me Again. After imputing my data, I will want to know if I have a good imputation, how to compare my data of study with my data of reference? and know the rate of imputation? and if there is a threshold ! Please thank you for helping me

[Kevin Blighe]

Hello again! As before, you may want to open a new question on that particular topic. This current question as been about problems running Beagle, and we have now solved that.

Just be sure to look at the Beagle manual before posting the new question, though.

[WouterDeCoster]

Kevin Blighe probably has the answer here already. Your example has "get=sortieMSL.recode.vcf". Kevin and the manual write "gt=sortieMSL.recode.vcf". As such, get is an unrecognized parameter and carefully looking in the manual would have solved this question.

[nour.hadjz]

i think i have to use "gt" beacause my file is a vcf file that contains genotypes. when i put gt=sortieMSL.recode.vcf" it tells me that my file "sortieMSL.recode.vcf" is not recognized.