pRESTO: a toolkit for processing high-throughput sequencing raw reads of lymphocyte receptor repertoires

Abstract

Driven by dramatic technological improvements, large-scale characterization of lymphocyte receptor repertoires via high-throughput sequencing is now feasible. Although promising, the high germline and somatic diversity, especially of B-cell immunoglobulin repertoires, presents challenges for analysis requiring the development of specialized computational pipelines. We developed the REpertoire Sequencing TOolkit (pRESTO) for processing reads from high-throughput lymphocyte receptor studies. pRESTO processes raw sequences to produce error-corrected, sorted and annotated sequence sets, along with a wealth of metrics at each step. The toolkit supports multiplexed primer pools, single- or paired-end reads and emerging technologies that use single-molecule identifiers. pRESTO has been tested on data generated from Roche and Illumina platforms. It has a built-in capacity to parallelize the work between available processors and is able to efficiently process millions of sequences generated by typical high-throughput projects.

Availability and implementation: pRESTO is freely available for academic use. The software package and detailed tutorials may be downloaded from http://clip.med.yale.edu/presto.

Fig. 1.

Example workflow diagram. Example workflows for single-end read sequencing protocols with sample barcoding (left) and paired-end read protocols with/without UID barcoding (right). Single sequence file inputs are shown with single arrowheads, and parallel processing of two paired-end read files are shown with a double arrowhead