SQUICH Data Processing

Created processed data from SQUICH experiments from the raw fastqs. This is data from the paper < insert official name of paper with a link to it >.

Files included

20180829_SQUISH14: Folder containing raw fastqs from SQUICH experiment

generate_counts.py: Script to process the raw reads in 20180829_SQUISH14 to get the code count table used for downstream SQUICH analysis

20180829_SQUISH14_collapsed_codes_align.csv: Parsed output file of generate_counts.py (created with the --one_tube flag)

20180829_SQUISH14_collapsed_codes_align_newcodelabel.csv: Parsed output file of generate_counts.py (created without the --one_tube flag)

all_libraries.csv: Contains metadata for each experiment corresponding to a fastq file in 20180829_SQUISH14

all_codes.csv: Contains concentrations of codes used in the experiments

all_degs.csv: Contains concentrations of degraders used in the experiments

all_spikes.csv: Contains concentrations of degraders used in the experiments

sq_utils.py: Contains some functions for working with sequence data

Running the Processing Script

To create the processed output for yourself using the script, clone this repo, cd inside it, and run the generate_counts.py script:

$ git clone https://github.com/salzmanlab/SQUICH_Data_Processing.git
$ cd SQUICH_Data_Processing
$ python generate_counts.py --one_tube

This script should take around 10 minutes to run.

This should create the file 20180829_SQUISH14_collapsed_codes_align.csv (this file is already included in this repository, so running the script should just overwrite that file in this directory with the same contents).

To create 20180829_SQUISH14_collapsed_codes_align_newcodelabel.csv (which separates out the code counts when all codes are added in the first round), run the following:

$ python generate_counts.py 

Again, this output file is already included in the repo so running this command should just overwrite it with the same file.