This repository contains all the datasets and analyses used to create the figures for PSERM paper. The numbered folders contain Jupyter Notebooks that were used to generate data and create the main figures. A conda environment is included which should easily allow for running any of the analyses. Most figures use non-default fonts (Myriad Pro and Courier New) and you will need to update fpath and aapath to the corresponding locations of the fonts saved on your machine.
The main code for analyzing the raw fastq files from the sequencer and creating the sequence:frequency csv files (ngs.py) and the code for computing all PSERMs and scoring (pserm.py) is included within this repository. ngs.py contains several classes used to compute and store the sequence and frequency data. These classes are used as input for the ngs_analysis object of pserm.py. To use ngs.py and pserm.py with these datasets, refer to the Jupyter Notebooks to see examples of how these objects are called and initiated.
The deep sequencing datasets used for this paper can all be found within the Input_Datasets folder under each of the respective projects. These datasets only show the mutated positions of the antibody and not the full antibody sequences.
The Project_PPMs and Project_PSSMs folders contain the PPMs or PSSMs computed within the notebooks. You can create them every time with ngs_analysis.generate_PSSM, however it's much quicker to save them and use the ngs_analysis.load_ppm or ngs_analysis.load_pserm functions.