PlasmidCoverage

Disclaimer

This script is no longer supported, if you are interested in the development of this pipeline check one of the following projects:

• FlowCraft
• pATLASflow

About

PlasmidUNCover.py is a script that is intended to allow users to map reads against a plasmid database (provided in indexes folder) . Then, the resulting accession numbers that have a percentage of covered base pairs higher than the specified cutoff value will be reported into a .json file which may be imported by Plasmid Atlas.

Other uses

It is also possible to map against any reference fasta, since it will automatically recognize the option provided and construct the required index files for that reference you want.

How to install

• First install bowtie2 (v2.2.9 or higher) and samtools (v.1.3.1 or higher).

Using pipy (recommended)

• pip3 install plasmiduncover

• Download indexes folder and uncompress it.

Using github release

• Second, download the latest release of this script (don't forget to download indexes folder).

• pip3 install -r requirements.txt

• Uncompress the indexes.tar.gz file.

Example run

If you installed using git clone

PlasmidUNCover.py -idx <path/to/indexes_folder> -r <path/to/reads_folder> -t 1 -o test -c 0.6

Note that each read or pair of reads should be inside its own folder within the path/to/reads_folder/. This allows the user to run multiple samples at once, since PlasmidUNCoverage.py will crawl this directory to search for directories with samples.

For advanced users, you may use PlasmidUNCoverage.py -h for additional options.

Dependencies

Note: ignore this if you have read the How to install section.

• plotly - pip install plotly
• termcolor
• bowtie2 (tested for version 2.2.9)
• samtools (tested for version 1.3.1)

You can now simply: pip3 install -r requirements.txt

Important note regarding the parsing of argument -r

You should provide the path to the directory containing the directories with the reads (.fastq files). For instance if you have something like ~/Reads/sample1/example.fastq and ~/Reads/sample2/example2.fastq, you should provide ~/Reads as the argument for -r option.

Options for PlasmidUNCoverage.py:

-p, --plasmid - Provide the path to the directory containing plasmid fastas.

-idx, --bowtie-index - Provide the path to bowtie index file

-r, --read - Provide the path to the directory containing reads fastas.

-t, --threads - Specify the number of threads to be used by bowtie2.

-2, --pair - Use this 
option if you have paired end reads. Paired end reads just have to be inside 
the same folder, without any other files inside it. Structure to read files 
should be something like: ./reads/read_folder/<with pairs inside it>.)

-k, --max_align - Specify the maximum number of alignments possible for each 
read. This option changes -k parameter of Bowtie2. By default this script will 
set -k to the number of fasta files in reference directory (e.g. if you have 3 
reference sequences the number of max_align allowed will automatically be set 
to 3). So, if you have a single multi-fasta -k will be set to 1.

-o, --output - Specify the output name you wish. There is no need of file 
extension.)

-c, --cutoff - Specify the cutoff for percentage of plasmid coverage that reads
 must have to be in the output. This should be a number between 0-1.')

---

- Accessory tools

plasmid_or_not.py

Currently this tool is separated from the main script (PlasmidCoverage_Sdb.py). This script is intended to separate reads from plasmids (or other type of sequences that have a database in fasta format) from reads for mixed libraries of chromosomal + plasmid reads.

Options for plasmid_or_not.py

-p, --plasmid - Provide the plasmid fastas

-r, --read - Provide the path to the directory containing reads fastas

-t, --threads - Specify the number of threads to be used by bowtie2

-o, --output - Specify the output name you wish. No need for file extension! 
Output file will be a fasta.

-unmap, --unmapped - By default this script attempts to save sequences 
available in the provided read files. If you want to save the reads that do not 
belong to plasmids, use this option.

diffs_json.py

Currently this tool is separated from the main script (PlasmidCoverage_Sdb.py). This script compares two json files with coverage percentage per gi (json files retrieved by PlasmidCoverage_Sdb.py.

Options for diffs_json.py

-i, --input_jsons - Provide the input json files of interest.

-c, --cutoff - Provide the cutoff value used for the minimum coverage allowed
 to be outputed to json files. This value should be equal in both files to compare.