install annovar and use it to annotate a vcf with hg19

gistfile1.eclass

wget http://www.openbioinformatics.org/annovar/download/annovar.latest.tar.gz.mirror
tar xzvf annovar.tar.gz
cd annovar

# download databases (goes to UCSC)
./annotate_variation.pl -buildver hg19 -downdb 1000g2010nov humandb
./annotate_variation.pl -buildver hg19 -downdb avsift humandb
./annotate_variation.pl -buildver hg19 -downdb refGene humandb
./annotate_variation.pl -buildver hg19 -downdb mce46way humandb/
./annotate_variation.pl -buildver hg19 -downdb snp131 humandb/
./annotate_variation.pl -buildver hg19 -downdb segdup humandb/
# more extensive gene db. see:
# http://www.openbioinformatics.org/annovar/annovar_gene.html 
# (Switching to UCSC Known Gene annotation or Ensembl Gene annotation)
./annotate_variation.pl --buildver hg19 --downdb knowngene humandb
./annotate_variation.pl --buildver hg19 --downdb ensgene humandb
# known variation
./annotate_variation.pl --buildver hg19 --downdb dgv humandb/
./annotate_variation.pl --buildver hg19 --downdb gwascatalog humandb/
# not available yet:
# perl ./annotate_variation.pl -buildver hg19 -downdb tfbs humandb/

./convert2annovar.pl -format vcf4 $IN > ${IN}.annovar

# see: http://www.openbioinformatics.org/annovar/annovar_faq.html#1000g_hg19

./annotate_variation.pl --filter --buildver hg19 --dbtype 1000g2010nov_all ${IN}.annovar humandb/
# iterative filtering the suffix hg19_ALL.sites... gets added. _filtered are the ones 
# that were not in the 1000genomes snps.
./annotate_variation.pl --filter --buildver hg19 --dbtype snp131  ${IN}.annovar.hg19_ALL.sites.2010_11_filtered humandb/

# now ${IN}.annovar.hg19_ALL.sites.2010_11_filtered.hg19_snp131_filtered
# contains the stuff not in dbsnp131 or in 1000 genomes.

# http://www.openbioinformatics.org/annovar/annovar_faq.html#vcf
# maybe can use this: http://biostar.stackexchange.com/questions/3432/1000g-and-dbsnp-build-132-in-ucsc-genome-browser/3436#3436
# for dbsnp 132

# TODO: http://www.openbioinformatics.org/annovar/annovar_faq.html#iupac
# IUPAC calls are excluded as bad input. convert these first

FILTERED=${IN}.annovar.hg19_ALL.sites.2010_11_filtered.hg19_snp131_filtered
# annotate remaining snps by proximity to gene.
./annotate_variation.pl --buildver hg19 --geneanno $FILTERED humandb/

# now ${FILTERED}.exonic_variant_function lists snps in genes
# and ${FILTERED}.variant_function annotates up/down stream genes and distances and if in UTR/intron

# generate a big spreadsheet of the annotations.
./auto_annovar.pl --buildver hg19  --ver1000g 1000g2010nov --verdbsnp 131 --genetype knowngene --outfile lung_auto LTRC_274462_lung_unique.annovar humandb