Bulk RNAseq Immune Clonotype Analysis (PubWeb)

Analysis of T-/B-cell clonotypes from bulk RNA sequencing data

Purpose

The goal of this workflow is to estimate the relative abundance of lymphocytes from different clonal lineages using bulk RNAseq data. The conceptual approach of the analysis is to reconstruct the TCR/BCR loci which were recombined in a unique way during the formation of each lineage.

Implementation

Input Data

Input data must be bulk RNAseq data in gzip-compressed FASTQ format. Data may be either single-end or paired-end.

To define the structure of the input, the user may specify either a filepath wildcard (e.g. path/to/reads/*_R{1,2}*.fastq.gz) or with a manifest file.

Manifest files must be formatted as a three-column CSV with the headers sample, fastq_1, and fastq_2.

To distinguish between paired- and single-end data, one (and only one) of the following parameters must be used:

• paired_reads
• paired_manifest
• single_reads
• single_manifest

Reference Genome

Reference sequences for any organism may be produced following the instructions laid out by the makers of TRUST4.

For convenience, reference sequences are provided in assets/TRUST4/ref/ for GRCm38, hg19, and hg38.

To select one of the pre-formatted reference sequences in that folder, the user may specify the parameter ref, which will reference the file assets/TRUST4/ref/{ref}_bcrtcr.fa. Alternatively, the user may specify any other reference sequence file with the param ref_fasta. Either ref or ref_fasta may be used, but not both.

Output Data

All output data will be written to the folder specified by the parameter output_folder