Annotating variants using Annovar question

[Phantom123]

I have vcf files which i generated using samtools. i wish to annotate variant information in these files using annovar. will the following work?

perl convert2annovar.pl in.vcf -format vcf4

perl annotate_variation.pl -buildver hg19 in.vcf humandb

in particular, will the above take care of the fact that inside the humandb folder (of annovar) i am finding hg18 related files?

Thanks for your help.

[Heisman]

If you want to keep all alternative alleles, first do this:

./convert2annovar.pl -format vcf4 --allallele variant_file > variant.annovarInput

Then I recommend doing everything in hg19 with the latest version, which has very fast performance with the "summarize_annovar.pl" script.

./summarize_annovar.pl --buildver hg19 --remove --verdbsnp 135 --ver1000g 1000g2012feb --outfile variant.output humandb/

This will annotate more fully than just the annotate_variation.pl script and is fast enough that it's worth using.

[HESmith]

Assuming that the vcf data are based on hg18, the positional information of the variants will not match hg19 annotations unless the reference genomes of hg18 and hg19 are identical. You'll need to convert your variant coordinates using liftover tables.

[Phantom123]

Dear All,
the first command that i gave in my original post as well as the first command given by heisman work fine on my system.

[These are:

perl convert2annovar.pl in.vcf -format vcf4

and

./convert2annovar.pl -format vcf4 --allallele variant_file > variant.annovarInput

]

But the second command given by me and also by Heisman are not working on my system.

For instance, i started off with a file called RVK127N.vcf

i then created a new file using this command:

perl convert2annovar.pl in.vcf -format vcf4 >RVK127N_out.vcf

i used the new file in the second command:

perl annotate_variation.pl -buildver hg19 RVK127N_out.vcf humandb

i then get the following message:
NOTICE: The --geneanno operation is set to ON by default
Error: The gene annotation database humandb/hg19_refGene.txt does not exist. Please use 'annotate_variation.pl --downdb refGene humandb -build hg19' to download the database.

i then type the following in the command line:

annotate_variation.pl --downdb refGene humandb -build hg19

but on pressing enter i get this message:

-bash: annotate_variation.pl: command not found

---
One other question: i have the entire ucsc hg19 reference genome in a file called genome.fa (downloaded independently from UCSC). do i use this genome.fa file in any of the annovar commands?

----
Thanks for your help.

[Heisman]

You do not need to use the genome.fa file anywhere.

As for:

annotate_variation.pl --downdb refGene humandb -build hg19

You need to use "perl annotate_variation.pl" or "./annotate_variation.pl"

Also, you want to use a command line more like this:

./annotate_variation.pl -downdb -webfrom annovar -buildver hg19 refGene humandb/

[Phantom123]

Much obliged to Heisman.