This small pipeline predicts the serotype of Salmonella samples from fastq files. It uses the tool SeqSero2 to do so (see: https://github.com/denglab/SeqSero2). It contains only two rules:
- Serotype prediction using SeqSero2 taking fastq files as input
- Creates a salmonella_multi_report.csv that collects all the results from all the samples run
Note: SeqSero2 is run in microassembly mode and, in this pipeline, it can only accept two separate fastq files (one for forward and one for reverse reads).
There are some parameters that can be changed. For instance, this pipeline automatically creates one new folder (or uses an existing one) called 'output/' in the current directory. You can change this behaviour by modifying the file config.yaml that is located in the 'config/' folder. Just go to the 'config:' section inside the config.yaml and change the "output_dir='output'" to "output_dir='your_output_dir'" where 'your_output_dir' can be any folder name or folder path you want.
Once you have done that, you can just run the pipeline by typing in your command line:
bash start_pipeline.sh -i <path_to_fastq_files> -o <path_to_desired_output_folder>
where the <path_to_fastq_files> is the input directory (-i) given either as an absolute path or as a path relative to your current directory. The same counts for your desired output directory (-o).
The pipeline also accepts other input that can be passed to snakemake. You can display the different options by typing:
bash juno-salmonella -h
or
bash juno-salmonella --help
The default is to run in the cluster ('bio' queue). If you want to change the queue, use --queue my_queue_name. If you want to run it locally you have to add the --local flag. You may also provide a specific number of --cores.