egustavsson / pipeline-ref-isoforms-ont

Pipeline for annotating genomes using long read transcriptomics data with stringtie and other tools

This snakemake pipeline creates GFF annotation from Oxford Nanopore cDNA or direct RNA reads. The cDNA reads are optionally processed by pychopper for trimming and orientation. The processed reads are mapped to the reference genome using minimap2, and then processed by stringtie in long read mode (with or withouth using a guide annotation) to generate the GFF annotation.

The annotation generated by the pipeline is compared to the existing annotation (if one can be found) using gffcompare and a report is genereated.

Getting Started

Input

• The input reads must be in fastq format.
• The input genome must be in fasta format.
• The existing annotations must be in GFF3 or GFF2 format.

Output

The main output files generated by the pipeline are the following:

• results/annotation/str_merged.gff - annotation in GFF2 format
• results/gffcompare/* - comparison to the existing annotation (if provided) by the gffcompare tool (documentation)
• results/gffcompare/str_gffcmp_report.pdf - visualisation of gffcompare results

Depedencies

• miniconda
• snakemake - easily installed via conda
• The rest of the dependencies are installed via conda.

Installation

Clone the pipeline and the pipeline toolset by issuing:

git clone --recursive https://github.com/nanoporetech/pipeline-ref-isoforms-ONT.git

Usage

Edit config.yml to set the input genome, input fastq and parameters, then issue:

snakemake --use-conda -j <num_cores> all