Preprocessing.py 文件源码

def PCR_preprocess(file_path, log_mode = False, pseudotime_mode = False, 
                   pcv_method = 'Rprincurve', anchor_gene = None,
                   exclude_marker_names = None):

    low_gene_fraction_max = 0.8

    data_tag, output_directory = create_output_directory(file_path)

    cell_IDs, cell_stages, data = get_PCR_or_RNASeq_data(file_path, pseudotime_mode)

    with open(file_path, 'r') as f:
        markers = np.loadtxt(f, dtype = str, delimiter = '\t', 
            skiprows = 1 if pseudotime_mode else 2, usecols = [0])
        markers.reshape(markers.size)

    if exclude_marker_names:
        indices = np.zeros(0, dtype = int)
        for name in exclude_marker_names:
            indices = np.append(indices, np.where(markers == name)[0])

        data = np.delete(data, indices, axis = 1)
        markers = np.delete(markers, indices)

    if pseudotime_mode:
        cell_stages = infer_pseudotime(data, output_directory, data_tag, pcv_method,
                                       anchor_gene, markers)

    condition = np.mean(data == 0, axis = 0) < low_gene_fraction_max
    data = np.compress(condition, data, 1)
    markers = np.compress(condition, markers)

    write_preprocessed_data(output_directory, cell_IDs, cell_stages, data, markers)

    return cell_IDs, data, markers, cell_stages.astype(float), data_tag, output_directory