- GAWN now requires blastplus 2.7.1+
- UTR region annotation has been turned off since v0.3
- The dependencies (mostly versions) have been updated
Developed by Eric Normandeau in Louis Bernatchez's laboratory with suggestions and important code contributions from Jérémy Le Luyer.
GAWN is a genome annotation pipeline that uses assembled transcriptome, either from the same species or from a related species, to create an evidence-based genome annotation. Its primary goal is to provide good enough genome annotation at a fraction of the time and effort required to run traditional genome annotation pipelines. It uses existing tools, such as GMAP, TransDecoder, blastx, the Swissprot database, etc. to produce the annotation. The result files are:
- A GFF3 annotation file
- A transcript annotation .tsv table
- A genome annotation .tsv table
The .tsv tables are formatted to maximize usability by non-specialized users.
This approach is especially useful to annotate genomes of species for which there is a good assembled transcriptome. It will also work when a good transcriptome is available for a related species. It provides only gene annotations for available transcripts. As such, it does not depend on ab initio gene prediction models.
During the analyses, the following steps are performed:
- Index the genome (
GMAP) - Annotate genes using available transcripts (
GMAP) - Annotate the transcripts (
blastxand the Swissprot database) - Produce a transcriptome annotation table (Python script)
- Produce a genome annotation table (Python script)
- TODO: add CpG island annotations
GAWN depends on different tools to annotate genomes. The requirements in terms of RAM, disk space, and time, is dependent on these tools. Here are example requirements for three different eukaryote genomes. The annotation was run on a Lenovo ThinkStation D20 with 8 Xeon CPUs (16 threads, 2.40GHz) on Linux Mint 17 (Ubuntu 16.04). All of these datasets, except Salvelinus fontinalis were run using the most recent genomes and transcriptomes available from Genbank.
| Genome | Size (Gbp) | RAM (GB) | Final disc space (GB) | Time (h) |
|---|---|---|---|---|
| Human genome | 3.29 | 16 | 37 | XX |
| Salvelinus fontinalis | 2.67 | 14.3 | 31.2 | XX |
| Danio rerio | 1.70 | XX | XX | XX |
| Drosophila melanogaster | 1.45 | 10.2 | 3.1 | 28 |
To use GAWN, you will need a local copy of its repository, which can be found here.
Different releases can be accessed here. We suggest using version 0.3.1 or a more recent version.
You will also need to have the following programs installed on your computer. The version numbers are the ones that have been tested. It is suggested that you use these or more recent versions.
- GNU Linux or OSX
- bash 4+
- python 2.7+ or 3.6+
- cufflinks v2.2.1+
- gmap (2017-10-12)
- wget 1.17.1
- gnu parallel 2017xxxx+
- blastplus utilities (blastx) 2.7.1+
- a local copy of the swissprot database (the .phr, .pin, .pnd, .pni ... files)
The relevant TransDecoder scripts are included with their license in
01_scripts/TransDecoder.
For each new project, get a new copy of the project's repository from the
sources listed in the Installation section and copy your data in the
03_data folder.
- Install dependencies
- Download GAWN (see Installation section above)
- Put your genome and transcriptome fasta files (uncompressed) in
03_data - Make a copy of
02_info/gawn_config.shand edit the parameters - Run the following command:
./gawn 02_info/MY_CONFIG_FILE.shOnce the pipeline has completed, all result files are found in the 05_results
folder.
- A valid gff3 annotation file
- A transcriptome annotation .tsv table
- A genome annotation .tsv table
A test dataset is available as a sister repository on GitHub.
Download the repository and then move the data in GAWN's 03_data folder.
Follow the normal pipeline procedure to analyse this small dataset. It should
run in one to ten minutes depending on your computer.
CC share-alike
GAWN by Eric Normandeau is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Based on a work at https://github.com/enormandeau/gawn.
