The pipeline gets samples in fasta format and provides whole genome alignment of desired samples on specified reference.
Obtained alignment could demonstrate interesting rearrangements on the chromosome level.
The resulting plot of alignment you can find in data_output folder.
Used tools:
git clone https://github.com/rutaolta/chromodifpipe.git
cd <pipeline_working_dir>
It is recommended to create a fresh conda environment using mamba or conda.
mamba env create --name chromodifpipe --file ./environment.yaml
# or:
# conda env create --name chromodifpipe --file ./environment.yaml
Activate conda environment with snakemake:
conda activate chromodifpipe
Before running the pipeline you should add whitelist of scaffolds you are interested in.
To check scaffold length please use following command.
That can be useful when choosing scaffolds length boundary to generate whitelists.
The generated reports will be put in data_input/reports folder.
snakemake -pr --use-conda --cores 1 scaffold_length
To generate whitelists of scaffolds please use following command. This step is required for alignment plot.
Whitelists would be generated for each sample using boundary.
Boundaries should be added into config/default.yaml parameter boundary.
The generated whitelists you can find in data_input/[name of your reference]/whitelists folder.
snakemake -pr --use-conda --cores 1 generate_whitelists
Scaffolds of sample will appear on plot in order that they appear in the .whitelist.
You could also add synonyms of scaffolds in the second column using Tab between columns.
The upper directory is named by the reference for you to be able to change the order of scaffolds if you will mind to use another reference.
There are 2 yeast samples in data_input/samples folder for pipeline test. For test data config/default.yaml has been modified as following:
-
samples_dirdefined as data_input/samples -
referencespecified as a cerevisiae -
speciesfor RepeatMasker specified as saccharomyces (The species name must be a valid NCBI Taxonomy Database species name and be contained in the RepeatMasker repeat database) -
boundaryfor whitelist generation is set to 1000000 -
filter_rangefor filtering "noise" is set to 1 -
plot_originalis true to generate also plot without filtering
To check the output you can run:
snakemake --cores 8 --configfile config/default.yaml --use-conda --profile profile/slurm/ --printshellcmds --latency-wait 60
and find results of test run in data_input folder
REMINDER: before run pipeline on your data your should specify settings in config/default.yaml corresponding to your samples. As an example you can have a look at test parameters above.
After all settings were specified you can run the pipeline with following command:
snakemake --cores 8 --configfile config/default.yaml --use-conda --profile profile/slurm/ --printshellcmds --latency-wait 60
Typically output plots with alignment should be filtered from "noise" that could interfere analysis results. Filter boundary could be defined in config/default.yaml parameter filter_range. Moreover if non-filtered plots look good and you only need to redraw filtered plots, you can skip drawing plot with originals to speed up. If you set parameter plot_original to false the pipe will redraw only plots based on filtered data.
And if you need to rerun plot step, for example if you added synonyms or changed the order of scaffolds/chromosomes in whitelist.txt, then you should remove or rename somehow following files with old results data_output/mavr/filtered_<name of your reference>.png.