TaMI, is a k-mer base variant caller that directly scan raw FASTQ files and looks for variants without alignement by using a dictionnary of mutated k-mers constructed for specific targets.
TaMI looks for all possible single nucleotide mutations (1-nt indels as well) only on the targeted regions. TaMI is very fast, a whole human exome can be scan within a minute in a single thread. TaMI memory load depend on the size of the indexed target regions.
- Clone tami repositiory
- run
maketo compile tami - place the
tamibinary somewhere accessible from your$PATH.
First you need to create a TAM files, wich is a dictionnary of mutated k-mers composed of all possible mutation in the targets (bed file).
tami build -r reference.fasta -o file.tam file.bed
Then you need to scan FASTQ files to look for mutations indexed in the TAM.
tami scan file.tam reads_1.fastq.gz reads_2.fastq.gz > output.vcf
Jérôme Audoux, Alexandre Soriano.