preprocess.py

"""
Preprocessing script that extracts information from experimental data produced by an imager run.

To use this file, first change the variable DATA_PATH to be the path to your data, then run
python preprocess.py
"""
import sys

import matplotlib as mpl

from tools.metadata import read_metadata
from transaction.extract_intensity import extract_intensity_all_files, extract_intensity_movies
from transaction.extract_peaks import extract_peaks_images, extract_peaks_movies
from transaction.fastq_tile_segmentation import fastq_tile_segmentation
from transaction.filter_fastq import filter_fastq
from transaction.match_tiles import match_tiles, match_tiles_movies
from transaction.summarize import create_summaries
from transaction.summarize_movie import summarize_movie

if __name__ == '__main__':
    # make matplotlib color palette more colorblind friendly
    mpl.rcParams['axes.prop_cycle'] = mpl.cycler(color=['#377eb8', '#ff7f00', '#4daf4a',
                                                        '#f781bf', '#a65628', '#984ea3',
                                                        '#999999', '#e41a1c', '#dede00'])

    if len(sys.argv) == 1:
        # Change this variable to be the path to your experimental directory
        DATA_PATH = "/Volumes/T7/imager_data/20220509_ISDV2_glu/"
        N_CORES = 8
    else:
        DATA_PATH = sys.argv[1]
        N_CORES = int(sys.argv[2])
    # read _metadata.csv, _config.csv which are located at DATA_PATH
    metadata = read_metadata(DATA_PATH)

    # sort fastq records by the sequencer tile
    fastq_tile_segmentation(metadata, n_cores=N_CORES)
    # remove fastq records that don't match the regex specified by SEQUENCER_INDEX
    filter_fastq(metadata, N_CORES)
    # detect peaks in the images
    extract_peaks_images(metadata, N_CORES)
    # extract_peaks_movies(metadata, N_CORES)
    # align the coordinates from the detected peaks with those the fastq-coords
    match_tiles(metadata, N_CORES)
    # match_tiles_movies(metadata, N_CORES)
    # calculate the intensity of the clusters
    extract_intensity_all_files(metadata, N_CORES)
    # extract_intensity_movies(metadata, N_CORES)
    # accumulate all cycles together
    create_summaries(metadata, N_CORES)
    # summarize_movie(metadata, N_CORES)