##Download
wget -O VCFCommandLineTools-1.0.0.jar https://github.com/christopher-gillies/VCFCommandLineTools/blob/master/release/VCFCommandLineTools-0.0.1.jar?raw=true
- Install java
- Download the latest jar file from release folder of this repository
set vcftools="c:\Users\cgillies\Downloads\VCFCommandLineTools-0.0.1.jar"
set vcf="Z:\path\all.chr.snp.indel.phased.vcf.gz"
java -jar %vcftools% --vcf %vcf% --command viewGenotypes --site 14:106337470 --site 14:106341031 --site 14:106341805
java -jar %vcftools% --vcf %vcf% --command viewGenotypes --site 14:106337470 --site 14:106341031 --site 14:106341805 --nucleotide
export vcfTools="java -jar /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export vcf=~/Documents/workspace-sts-3.6.1.RELEASE/vsearchDataDirectory/reheader.neptune.all.snp.sorted.all.indels.eff.dbsnp.b138.esp.1kgp1.dbnsfp.2.5.hgmd.LowQC.0.7.targeted.vcf.gz
$vcfTools --vcf $vcf --command viewGenotypes --site 17:48156000-48157000 --site 17:48158680
SAMPLE_ID 17:48156449:A:G 17:48156557:G:A 17:48158680:A:G
SAMPLE_1 0 0 0
SAMPLE_2 0 0 0
SAMPLE_3 0 0 0
SAMPLE_4 0 0 0
$vcfTools --vcf $vcf --command viewGenotypes --site 17:48156000-48157000 --site 17:48158680 --nucleotide
SAMPLE_ID 17:48156449:A:G 17:48156557:G:A 17:48158680:A:G
SAMPLE_1 A/A G/G A/A
SAMPLE_2 A/A G/G A/A
SAMPLE_3 A/A G/G A/A
SAMPLE_4 A/A G/G A/A export vcfTools="java -jar /home/cgillies/programs/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export VCF=/kidneyomics/1000G_Phase3/ALL.chr1.phase3_shapeit2_mvncall_integrated_v5a.20130502.genotypes.vcf.gz
$vcfTools --command findOverlappingSamplesFromList --vcf $VCF --infile /kidneyomics/NEPTUNE_FLUIDIGM/1000G.samples.ids --outfile /kidneyomics/NEPTUNE_FLUIDIGM/1000G.samples.ids.phase3.overlap.txt
cat /kidneyomics/NEPTUNE_FLUIDIGM/1000G.samples.ids.phase3.overlap.txt
HG00100
HG00138
HG00160
HG00233
HG00246
...
This examples requires that you specify an input file with a list of variants. In this file, each line has the following format: "chr:pos:ref:alt". There should be no header line. The output option is optional.
export vcfTools="java -jar /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export VCF=/Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/FluidigmReferences/ExAC.r0.3.sites.vep.prefix.decomposed.normalized.pass.vcf.gz
export variants=/Users/cgillies/Google\ Drive/Fluidigm_Pipeline/PAPER_ANALYSIS/exome_chip_sensitivity/singleton_variants.txt
export out=/Users/cgillies/Google\ Drive/Fluidigm_Pipeline/PAPER_ANALYSIS/exome_chip_sensitivity/allele_frequency.txt
$vcfTools --command viewInfo --vcf $VCF --info EXAC_AC_Adj --info EXAC_AN_Adj -info EXAC_AF --infile "$variants" --outfile "$out"
export vcfTools="java -jar /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export REF="/Users/cgillies/Google Drive/1_7_2016_Megachip/chr20.fa.gz"
export MANIFEST="/Users/cgillies/Google Drive/1_7_2016_Megachip/MEGA_Consortium_15063755_B2.chr20.csv"
export OUT="/Users/cgillies/Google Drive/1_7_2016_Megachip/chr20.sites.vcf"
$vcfTools --command makeVcfFromManifest --outfile "$OUT" --ref "$REF" --manifest "$MANIFEST"
export vcfTools="java -jar /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export REF="/Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/20.fa.gz"
export MANIFEST="/Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/manifest.test.chr20.csv"
export OUT="/Users/cgillies/Google Drive/1_7_2016_Megachip/gt.vcf.gz"
export IN="/Users/cgillies/Google Drive/1_7_2016_Megachip/sampleList.txt"
vi "$IN"
HG00100 /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/HG00100.chr20.txt
HG00138 /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/HG00138.chr20.txt
HG00160 /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/HG00160.chr20.txt
$vcfTools --command makeVcfFromReports --outfile "$OUT" --ref "$REF" --manifest "$MANIFEST" --infile "$IN"
export vcfTools="java -jar /home/cgillies/programs/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export VCF1="/home/cgillies/programs/VCFCommandLineTools/src/test/resources/ALL.chip.omni_broad_sanger_combined.20140818.snps.genotypes.chr20.subset.relabel.vcf.gz"
export VCF2="/home/cgillies/programs/VCFCommandLineTools/src/test/resources/ALL.chip.omni_broad_sanger_combined.20140818.snps.genotypes.chr20.subset.vcf.gz"
export OUT="/tmp/test.vcf.gz"
$vcfTools --command mergeVcfColumns --outfile "$OUT" --vcf "$VCF1" --vcf "$VCF2"
Example ld-prune (Note: this is setup for unphased markers and LD is estimated with Pearson correlation)
export vcfTools="java -jar /home/cgillies/programs/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export VCF1="/home/cgillies/programs/VCFCommandLineTools/src/test/resources/ALL.chip.omni_broad_sanger_combined.20140818.snps.genotypes.chr20.subset.relabel.vcf.gz"
export OUT="/tmp/test.vcf.gz"
$vcfTools --command filter --outfile "$OUT" --vcf "$VCF1" --maxLd 0.2 --windowSizeKb 1000 --minAc 2 --excludeChr Y --excludeChr X --excludeChr MT
Example filter by hwe across all population (remove if the excat test HWE p-value is less than 0.01)
export vcfTools="java -jar /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export VCF1="/Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/ALL.chip.omni_broad_sanger_combined.20140818.snps.genotypes.chr20.subset.vcf.gz"
export OUT="/tmp/test.vcf.gz"
$vcfTools --command filter --outfile "$OUT" --vcf "$VCF1" --hwe 0.01 --minAc 0
Example filter by hwe across each population (remove if the excat test HWE p-value is less than 0.01 in any population)
The population information is in a tab delimited infile, where the first row is a header line. The idCol option specifies the column of the sample id, and the popCol option specifies the population column
export vcfTools="java -jar /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export VCF1="/Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/ALL.chip.omni_broad_sanger_combined.20140818.snps.genotypes.chr20.subset.vcf.gz"
export ANCESTRY="/Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/omni_samples.20141118.panel"
export OUT="/tmp/test.vcf.gz"
$vcfTools --command filter --outfile "$OUT" --vcf "$VCF1" --hwe 0.01 --minAc 0 --infile $ANCESTRY --idCol sample --popCol pop
The --filterString option must return a R logical value. That is the last line of the script should be a logical value. Sites are kept if the result is TRUE otherwise it will filter the site out
- chr -- The chromosome
- start -- The start position
- stop -- The stop position
- id -- The variant id
- ref -- The reference allele
- alt -- The first alternative allele
- qual -- The quality score
- filters -- The filter information for the variant
- info -- A list() object representing the info field (only those annotated in the VCF header). For example, if you have an AC info item it could be accessed info[['AC']] in the R script.
- samples -- a list() of sample ids
- gtInfo -- a list() of lists() one for each sample. This can be used to pull of genotype specific format fields. For example to get a list of genotypes from all samples you could write
gts = sapply(gtInfo,FUN=function(x){ x[['GT']] });
- Note that the GT field is encoded as 0 for HOM_REF, 1 For HET, and 2 HOM_ALT.
export vcfTools="java -jar /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export VCF1="/Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/ALL.chip.omni_broad_sanger_combined.20140818.snps.genotypes.chr20.subset.vcf.gz"
export OUT="/tmp/test.vcf.gz"
$vcfTools --command filter --outfile "$OUT" --vcf "$VCF1" --minAc 0 --filterString "gts = sapply(gtInfo,FUN=function(x){ x[['GT']] }); sum( gts == 1 ) == 11;"
export vcfTools="java -jar /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export VCF1="/Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/ALL.chip.omni_broad_sanger_combined.20140818.snps.genotypes.chr20.subset.vcf.gz"
export OUT="/tmp/test.vcf.gz"
$vcfTools --command filter --outfile "$OUT" --vcf "$VCF1" --minAc 0 --hwe 0.1 --maxLd 0.1 --filterString "gts = sapply(gtInfo,FUN=function(x){ x[['GT']] }); sum( gts == 1 ) == 11;"
R filter using GCScore field from VCF. Only keep variants if the mean GCScore across all non-zero subjects is greater than 0.7
export vcfTools="java -jar /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export VCF1="/Users/cgillies/Google Drive/1_7_2016_Megachip/chr20.merged.reports.rc.dict.sorted.filtered.1000G.vcf.gz"
export OUT="/tmp/test.vcf.gz"
$vcfTools --command filter --outfile "$OUT" --vcf "$VCF1" --minAc 0 --filterString "gc = sapply(gtInfo,FUN=function(x){ x[['GCScore']] }); mean(gc,an.rm=T) > 0.7"
R filter using GCScore field from VCF. Only keep variants if the mean GCScore is greater than the mean non-zero GTScore
export vcfTools="java -jar /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export VCF1="/Users/cgillies/Google Drive/1_7_2016_Megachip/chr20.merged.reports.rc.dict.sorted.filtered.1000G.vcf.gz"
export OUT="/tmp/test.vcf.gz"
$vcfTools --command filter --outfile "$OUT" --vcf "$VCF1" --minAc 0 --filterString "gc = sapply(gtInfo,FUN=function(x){ x[['GCScore']] }); gt = sapply(gtInfo,FUN=function(x){ x[['GTScore']] }); mean(gc,an.rm=T) > mean(gt[gt != 0],na.rm=T)"
R filter using info field. Find the variant with IlluminaId == '20:10036280-GA-0_T_F_2299624158' and print the IlluminaId for every variant
export vcfTools="java -jar /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export VCF1="/Users/cgillies/Google Drive/1_7_2016_Megachip/chr20.sites.vcf"
export OUT="/tmp/test.vcf.gz"
$vcfTools --command filter --outfile "$OUT" --vcf "$VCF1" --minAc 0 --filterString "print(info[['IlluminaId']]);info[['IlluminaId']] == '20:10036280-GA-0_T_F_2299624158'"
The first --vcf is the truth vcf The second --vcf is the test vcf The first --sample is the truth sample id The second -sample is the test sample id
export vcfTools="java -jar /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export VCF1="/Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/ALL.chip.omni_broad_sanger_combined.20140818.snps.genotypes.chr20.subset.vcf.gz"
$vcfTools --command concordance --vcf "$VCF1" --vcf "$VCF1" --minAc 10 --sample HG00100 --sample HG00100
Truth VCF: /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/ALL.chip.omni_broad_sanger_combined.20140818.snps.genotypes.chr20.subset.vcf.gz
Test VCF: /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/ALL.chip.omni_broad_sanger_combined.20140818.snps.genotypes.chr20.subset.vcf.gz
Truth Sample: HG00100
Test Sample: HG00100
True Positives: 16554
False Positives: 0
True Negatives: 12676
False Negatives: 0
Sensitivity: 1.0
Specificity: 1.0
False Discovery Rate: 0.0
export vcfTools="java -jar /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/release/VCFCommandLineTools-0.0.1.jar"
export VCF1="/Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/ALL.chip.omni_broad_sanger_combined.20140818.snps.genotypes.chr20.subset.vcf.gz"
$vcfTools --command concordance --vcf "$VCF1" --vcf "$VCF1" --minAc 10 --sample HG00100 --sample HG00100 --outfile "/tmp/test.log" --snpsOnly
Truth VCF: /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/ALL.chip.omni_broad_sanger_combined.20140818.snps.genotypes.chr20.subset.vcf.gz
Test VCF: /Users/cgillies/Documents/workspace-sts-3.6.1.RELEASE/VCFCommandLineTools/src/test/resources/ALL.chip.omni_broad_sanger_combined.20140818.snps.genotypes.chr20.subset.vcf.gz
Truth Sample: HG00100
Test Sample: HG00100
True Positives: 16554
False Positives: 0
True Negatives: 12676
False Negatives: 0
Sensitivity: 1.0
Specificity: 1.0
False Discovery Rate: 0.0
head /tmp/test.log
20:48270776:C:T 0 20:48270776:C:T 0 TN
20:44916134:C:T 1 20:44916134:C:T 1 TP
20:12794193:G:A 2 20:12794193:G:A 2 TP
20:8662410:A:G 1 20:8662410:A:G 1 TP
20:52437475:T:C 1 20:52437475:T:C 1 TP
20:51727616:C:T 1 20:51727616:C:T 1 TP
20:23592714:T:C 0 20:23592714:T:C 0 TN
20:46483476:T:C 0 20:46483476:T:C 0 TN
20:39103371:G:A 1 20:39103371:G:A 1 TP
20:38560263:A:G 1 20:38560263:A:G 1 TP
The columns are [KEY for Truth] [GT for Truth] [Key for Test] [GT for Test]
$vcfTools --help
_ _ _ ___
\ / / |_ / _ ._ _ ._ _ _. ._ _| | o ._ _ | _ _ | _
\/ \_ | \_ (_) | | | | | | (_| | | (_| |_ | | | (/_ | (_) (_) | _>
usage: VCFCommandLineTools
--command <command> The command you would like to perform:
findOverlap, selectSites,
viewGenotypes, viewInfo,
findOverlappingSamplesFromList,
makeVcfFromManifest,
makeVcfFromReports. findOverlap
requires you to input at least two vcf
files and the program will find the
samples biallelic sites in both vcf
files. selectSites will select
biallelic sites from the file that you
specify with format chr:pos:ref:alt
for each variant. viewGenotypes will
display the genotypes for sites of
interest. viewInfo will display a
variants information from info field.
makeVcfFromManifest takes an input of
a illumina manifest file and creates a
vcf sites file for sites that have a
reference allele. makeVcfFromReports
creates a vcf from illumina standard
report files. mergeVcfColumns merges
two vcf files (only biallelic sites;
no duplicate sample ids). filter --
remove variants from vcf, can be used
to ld-prune. concordance -- between two
samples from different vcfs
--excludeChr <excludeChr> please specify a chr to exclude
--filterString <filterString> The filter string to appy to the
filter command. Each variants
genotypes will be put into a variable
gtInfo. This is a list that acts much
like a hash table. Any valid R code
can be used, however, it MUST return a
LOGICAL R value.
--help Print the help message
--hwe <hwe> The Hardy-Weinberg p-value threshold
(Exact test); used in filtering
--idCol <idCol> The identity column to pull out of the
infile. Requires a header
--infile <infile> The file to read in
--info <info> The info to select out of vcf
--manifest <manifest> please specify a illumina manifest
file
--maxLd <maxLd> The maximum allowable pairwise ld when
filtering variants
--minAc <minAc> The minimum allele count for a variant
to be considered
--nucleotide Show nucleotides instead of numeric
genotype
--outfile <outfile> The file to write out to
--popCol <popCol> The population column to pull out of
the infile. Requires a header
--ref <ref> please specify a reference sequence
--sample <sample> specify a sample id
--site <site> please specify a site
--vcf <vcf> a vcf file
--windowSizeKb <windowSizeKb> The window size in kilobases when
filtering variants