Usage

Installation

To use CCTK, first install it. The easiest installation method is using conda. CCTK is currently available on my personal Anaconda channel. In the future I intend to host it through Bioconda.

Conda - Recommended

N.B. Some dependencies of CCTK are distributed through the bioconda and conda-forge channels. If you do not have those in your conda config you can add them as follows:

conda config --append channels conda-forge
conda config --append channels bioconda

You can then create a conda environment and install CCTK and all dependencies into it with the following:

conda create -n cctk cctk
conda activate cctk

If you do not wish to modify your conda config you will need to specify the above channels in your installation command.

conda create -n cctk -c conda-forge -c bioconda cctk

Conda has the benefit of handling the installation of the correct version of all depencies and adds the cctk (and dependency) executables to your PATH, which makes usage simpler.

Git

You can also install dependencies seperately and download CCTK from github.

To download CCTK using git:

$ git clone https://github.com/Alan-Collins/CRISPR_comparison_toolkit.git

You can find the cctk executable at CRISPR_comparison_toolkit/cctk

Dependencies

• python3 (Tested with version >= 3.8)

  • Matplotlib (Tested with version 3.5.0)

  • DendroPy (Tested with version 4.5.2)

  • NumPy (Tested with version 1.21.2)

• MinCED

• BLAST+ (tested with version >= 2.9.0)

Quick run

CCTK includes a quickrun command to run some CCTK tools with default settings to allow you to get a quick look at your data. cctk quickrun identifies CRISPRs using CCTK Minced and then visualizes all identified groups of related arrays (groups that contain between 3 and 15 arrays by default) using CCTK CRISPRdiff and CCTK CRISPRtree.

cctk quickrun is a good way to check if your dataset contains any related arrays and to get a sense of their relationships. If you would like to see what a more involved analysis using CCTK would look like then have a look at the tutorial page.

Example Workflow

An example workflow of using all of the tools in CCTK is given in the tutorial page.

Usage Help

During all examples, we will use CCTK as if it were installed using conda and is in our PATH.

All tools available in CCTK can be accessed through the cctk executable. Each tool can be executed by calling the corresponding command within cctk. A list of available commands can be found in the help message returned by the cctk executable when run with the -h or --help options.

(cctk) $ cctk -h
usage: cctk [-h] [--version]

optional arguments:
  -h, --help  show this help message and exit
  --version   show program's version number and exit

Available commands in the CRISPR comparison toolkit:

  Call any command followed by -h or --help to show help for that command

  Find CRISPR arrays in assemblies:
    blast        find CRISPR arrays with user-provided repeat(s) using BLASTn
    minced       find CRISPR arrays using minced

  Analyze the differences between CRISPR arrays:
    crisprdiff   produce a CRISPRdiff plot comparing CRISPR arrays
    crisprtree   perform a maximum parsimony analysis on CRISPR arrays
    constrain    predict array relationships constrained by a tree ## TO ADD ##
    network      produce a network representation of spacer sharing among arrays

  Other:
    evolve       perform in silico evolution of CRISPR arrays
    spacerblast  BLAST spacers against a BLASTdb, process output & check for PAMs
    quickrun     Run a default CCTK pipeline

Usage of each command can be found by calling that command with -h or --help. e.g. cctk blast -h

Details of the specific usage for each tool in CCTK can be found in the CCTK Tools section.

Adapting your data from a non-CCTK pipeline

CCTK is primarily built around two files:

1. a fasta file of your spacer sequences, and

2. Array_IDs.txt

If you identified CRISPR arrays using a method other than CCTK, then in order to perform analyses using CCTK you will need to generate one or both of these two files. For the tools CCTK Network, CCTK CRISPRdiff, CCTK CRISPRtree, and CCTK Constrain, you only need the Array_IDs.txt file. For CCTK Spacerblast, you will need a fasta file of your spacers.

How you can convert your data to these CCTK data formats will depend on the method you used to identify CRISPR arrays. However, CCTK does provide one simple way to generate all of the CCTK files if you already have CRISPR spacers in fasta format. Specifically, CCTK Minced and CCTK Blast have an option that allows Appending to an existing dataset. You can use that option using the following steps to run CCTK Minced or CCTK Blast:

N.B., We recommend using CCTK Blast for this, assuming you know the sequence of the CRISPR repeats in your assemblies. Minced can make mistakes about repeat boundaries and can identify false-positive arrays. If you know your repeat sequences, CCTK Blast will produce a better output.

1. Make an output directory to store the CCTK output files. (If running CCTK Minced, make another directory inside of this one called “PROCESSED”)

2. Copy your CRISPR spacers fasta file into the newly created directory (the PROCESSED directory if using CCTK Minced )

3. Run CCTK Minced or CCTK Blast with --append and whichever other options you wish

You should now have a directory containing all of the normal CCTK output files. The spacer fasta file you provided was read by CCTK and used during the process of identifying CRISPR arrays. In addtion, any identified CRISPR spacers that were not in the original set you provided will have been added to the end of the file. If you look at the produced files, you should see that the spacers you provided and their fasta headers have been used by CCTK.