decombinator is a fast and efficient tool for the analysis of T-cell receptor (TCR) repertoire sequences produced by deep sequencing. It accepts TCR repertoire sequencing (TCRseq) FASTQ data and returns AIRR compliant files detailing TCR V(D)J recombination and counts.
TCRseq offers a powerful means to investigate biological samples to observe sequence distributions. However current high-throughput sequencing (HTS) technologies can produce large amounts of raw data which will also unavoidably contain errors relative to the original input molecules.
decombinator addresses the problem of large datasets through speed - employing a rapid and highly efficient string matching algorithm to search the FASTQ files produced by HTS machines for rearranged TCR sequences. The central algorithm searches for 'tag' sequences, the presence of which uniquely indicates the inclusion of particular V or J genes in a recombination. If V and J tags are found, decombinator can then deduce where the ends of the germline V and J gene sections are (i.e. how much nucleotide removal occurred during V(D)J recombination), and what nucleotide sequence (the 'insert sequence') remains between the two.
These five pieces of information - the V and J genes used, how many deletions each had and the insert sequence - contain all of the information required to reconstruct the whole TCR nucleotide sequence, in a more readily stored and analysed way.
The pipeline also handles FASTQ data with many reads of the same molecule due to over-amplification. It does this by using unique molecular identifiers to \"collapse\" down duplicate reads.
Running decombinator is easy. All that is required is HTS read data and a few arguments specifying the chemistry of your prepared library:
decombinator pipeline -in XXXX.fq -c b -br R2 -bl 42 -ol M13