Convert raw sequencing files into library counts using bowtie alignment
The script used for this step is fastqgz_to_counts.py, and requires a reference fasta file of the library used in the experiment. Indices for several published libraries are included here, and more can be generated upon request.
Generate sgRNA phenotype scores, gene-level scores, and gene-level p-values
This step relies on the script process_experiments.py. This script requires you to first fill out a configuration file which allows you to:
This script also generates a set of standard graphs using screen_analysis.py
[Optional] Generate graphs interactively using screen_analysis.py
(ScreenProcessing no longer uses Bowtie to align sequencing reads; if you want to use or fork from this functionality use an earlier version of the program)
A requirements.txt file has been added, as there may be issues with some current packages. This may not be the most recent functional version - testing is in progress. This file should be used to create a virtual environment.
Alternatively, a Singularity Definition file has been added, intended to be used to create a Singularity container that has the correct functional versions of dependencies. Here is how to create a container:
singularity build ScreenProcessing.sif ScreenProcessing.def
A PDF slideshow with a step-by-step tutorial of screen analysis using the data files included in the Demo folder can found here: ScreenProcessing Demo
The demo files represent a tiny slice of the full sequencing dataset to speed up the download and demo scripts. The full complement of sequencing data used for the cell growth and cholera toxin sensitivity CRISPRi screens published in Gilbert and Horlbeck et al., Cell 2014 can be accessed here: data link