This script processes raw data from a sensor based base editing screen into a mapping from sgRNAs to editing outcomes. In particular, the script uses raw, paired-end read data to classify the editing events that resulted in the observed target sites. It is assumes that the sensor screen has the following structure:
left-scaffold --- sgrna --- middle-scaffold --- target-site --- right-scaffold
and that the sgRNA region remains unedited during the screen. Reads with edited sgRNA regions are thrown out as errors.
To run the script, first compile the program with lein using
lein with-profile prod uberjar
Then, run the following command to analyze the raw read data.
$ java -jar analysis.jar
Usage: java -jar analyze-fastqs rep1 ... repN [options]
Options:
-o, --output FILE Output file (REQUIRED).
-w, --whitelist WHITELIST Whitelist file for screen (REQUIRED).
-c, --config CONFIG Configuration file describing sensor and whitelist structure (REQUIRED).
-f, --format OUTPUT_FORMAT TARGET Output either only target cytosine or all cytosines.
-h, --help
Expects FASTq files consisting of the paired end reads
for each replicate.