How to run a Gemini analysis on clia1

runGemini.sh

#VEP is on clia1 here - /opt/gms/vep/variant_effect_predictor

#An example that I used to run it was,
#VEP is installed here /opt/gms/vep/variant_effect_predictor
#PERL5LIB=$PERL5LIB:/opt/gms/vep/ perl variant_effect_predictor.pl -i example_GRCh37.vcf --cache --dir_cache /opt/gms/vep/

#another example with required VEP annotations for GEMINI
PERL5LIB=$PERL5LIB:/opt/gms/vep/ perl /opt/gms/vep/variant_effect_predictor/variant_effect_predictor.pl -i test.vcf \
--cache --sift b --polyphen b --symbol --numbers --biotype --total_length \
-o snvs.detailed.vep.vcf --vcf --fields Consequence,Codons,Amino_acids,Gene,SYMBOL,Feature,EXON,PolyPhen,SIFT,Protein_position,BIOTYPE \
--cache --dir_cache /opt/gms/vep/ --assembly GRCh37 --offline --stats_file test.html --fork 20

#load the VCF into Gemini
#Gemini is installed here - /opt/gms/gemini/bin/gemini
PATH=$PATH:/opt/gms/gemini/bin gemini load -v snvs.detailed.vep.vcf -t VEP GTB08.db --cores 30

#Gemini query
/opt/gms/gemini/bin/gemini query --header -q \
"SELECT v.chrom,v.start,v.end,v.qual,v.filter,v.num_het,v.aaf,v.gene,v.cosmic_ids,v.biotype,v.impact,v.impact_severity,v.aa_change,v.rs_ids \
FROM variants v, gene_summary g WHERE v.gene=g.gene and v.is_coding==1 and v.in_1kg == 0 and v.in_esp==0 and v.impact!='synonymous_coding' AND \
v.impact!='intron' AND v.impact!='downstream' AND v.impact!='upstream' and g.in_cosmic_census==1" GTB08.db

#query with clinvar
/opt/gms/gemini/bin/gemini query --header -q SELECT v.chrom,v.start,v.end,v.qual,v.filter,v.num_het,v.aaf,v.gene,v.cosmic_ids,v.biotype,v.impact,v.impact_severity,v.aa_change,v.rs_ids,v.in_omim,v.clinvar_disease_name FROM variants v, gene_summary g WHERE v.gene=g.gene and v.is_coding==1 and v.in_1kg == 0 and v.in_esp==0 and v.i
mpact!='synonymous_coding' AND v.impact!='intron' AND v.impact!='downstream' AND v.impact!='upstream' and g.in_cosmic_census==1 GTB08.db

A newer version of this script is on the cgs_scripts repository here - gatoravi/nowarranty@323b0c1