snpeffToMaf

SnpEff is more tiny than VEP, and more HGVS than ANNOVAR
e.g, java -jar snpEff.jar hg19 yourSampleName.normalize.vcf -hgvs1LetterAa -canon > yourSampleName.snpeff.vcf

🔥 support complex variant(greater than one base)

a test environment

• samtools 1.9 and bcftools 1.6
• SnpEff 4.3t
• R 3.6.3

usage example

1. in shell
   perl snpeffToMaf.pl yourSampleName.snpeff.vcf 100 0.05
   100 for minimum depth 0.05 for minimum allele frequency
   warn: you may reformat with GT:DP:AD if vcf created by other variant callers. https://samtools.github.io/bcftools/howtos/query.html + https://samtools.github.io/bcftools/bcftools.html#expressions

2. in shell
   cat *maf | awk '!/Hugo_Symbol/ || NR==1' > all.maf
   concatenate all of yourSampleName.maf, but not necessary

3. in R
   library(maftools)
syn <- c("synonymous_variant","start_retained","stop_retained_variant")
df <- data.table::fread("all.maf")
vc <- names(table(df$Variant_Classification))
nonSyn <- setdiff(vc,syn)
colors <- rainbow(length(nonSyn))
names(colors) <- nonSyn
maf <- read.maf("all.maf", vc_nonSyn = nonSyn)
plotmafSummary(maf, rmOutlier = TRUE, addStat = 'median', dashboard = TRUE, titvRaw = FALSE, color = colors)

screenshot

issues