Acanthamoeba castellanii genome annotation
cmdoret, 20190905
This pipeline allows to reproduce the automatic annotation procedure used for the A. castellanii genome assembly. Most of the work is done using funannotate. Each step of the pipeline is run inside a singularity container. The only dependencies are python>=3.7, funannotate, snakemake and conda.
The associated Zenodo record holds a frozen copy of the repository as well as the input and output data. The input data (genome assemblies used in the publication) are automatically downloaded from Zenodo upon execution of the pipeline.
Configuration
There are 3 configuration files:
config.yaml: General pipeline parameters as well as path to the output and temporary folders.samples.tsv: Assemblies to annotate.units.tsv: Describes input reads files to use for evidence during annotation.
Installation
Funannotate must be installed in the current conda environment and setup according to the official instructions (including setting the $FUNANNOTATE_DB variable). Snakemake >=5.5 is also required to run the pipeline.
Eggnog-mapper and interpsoscan should also be available as they will be called by funannotate to improve the annotations.
Usage
If you funannotate is available in your cuurrent conda environment, you can the pipeline with the following command:
snakemake --use-conda -j4Description
The pipeline works as follows:
- Download the assemblies from zenodo and rnaseq reads from SRA
- Clean the input assembly (rename headers, sort and filter scaffolds)
- Soft mask repeats from the assembly
- Use RNAseq data to predict genes with AUGUSTUS
- Use remote services for functional annotations (Eggnog-mapper, interproscan, phobius)
- Combine functional annotations from the different sources.
