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-nanopore-ref-isoforms.git

Usage

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

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

Results

Performance on a Drosphilia melanogaster with cDNA-PCR Sequencing (SQK-PCS109) data

The data used for benchmarking generated with the Oxford Nanopore Technologies SQK-PCS109 protocol can be downloaded from ENA.

Performance with guide annotation

Run ./run_evaluation_dmel.sh to reproduce these results.

See this folder for raw gffcompare output and the full PDF report.

Performance without guide annotation

See this folder for raw gffcompare output and the full PDF report.

HMM logos of alignment contexts

HMM logos are generated from the context of the start and end positions of the read alignments and saved under alignments/context*.png. Some examples are shown below.

The logo of the context of aligment start positions of a random sample of reads. Note the pattern reminescent of a TATA-box at the beggining.

The logo of the context of aligment end positions of a random sample of reads. Note that the region is AT rich, which is a characteristic of polyadenylation signals.

Licence and Copyright

(c) 2020 Oxford Nanopore Technologies Ltd.

This Source Code Form is subject to the terms of the Mozilla Public License, v. 2.0. If a copy of the MPL was not distributed with this file, You can obtain one at http://mozilla.org/MPL/2.0/.

FAQs and tips

• The GFF2 files can be visualised using IGV.

References and Supporting Information

See the post announcing the tool in the Nanopore Community here.

Research Release

Research releases are provided as technology demonstrators to provide early access to features or stimulate Community development of tools. Support for this software will be minimal and is only provided directly by the developers. Feature requests, improvements, and discussions are welcome and can be implemented by forking and pull requests. However much as we would like to rectify every issue and piece of feedback users may have, the developers may have limited resource for support of this software. Research releases may be unstable and subject to rapid iteration by Oxford Nanopore Technologies.