| |
| |
| from prepare_for_featurization.prepare_for_featurization import main as split_n_pick |
| from generate_inputs.generate_inputs import samples_from_file |
| from utils.utils import make_sub_folder |
|
|
| def main(num_splits, |
| max_len, |
| seed_folder, |
| trees_folder, |
| cherries_folder): |
| split_file_lst = ['pfams_in_OOD_valid.tsv'] + [f'pfams_in_split{i}.tsv' for i in range(num_splits)] |
| for pfam_filename in split_file_lst: |
| splitname = pfam_filename.replace('.tsv','').split('_')[-1] |
| cherries_dset_prefix = f'CHERRIES_{splitname}' |
| |
| with open(f'{pfam_filename}','r') as f: |
| pfams_in_split = [line.strip() for line in f] |
| |
| file_lst = [f'{cherries_folder}/{pf}_cherries.tsv' |
| for pf in pfams_in_split] |
| |
| for suffix in ['full_length', 'summarized_counts', 'all_metadata']: |
| this_split_cherries_folder = f'{cherries_dset_prefix}_{suffix}' |
| make_sub_folder(in_dir = '.', sub_folder=this_split_cherries_folder) |
| |
| for i in range(len(pfams_in_split)): |
| if i % 10 == 0: |
| print(f'{i}/{len(pfams_in_split)}') |
| |
| pfam = pfams_in_split[i] |
| file_of_cherries = file_lst[i] |
| |
| samples_from_file(pfam = pfam, |
| seed_folder = seed_folder, |
| trees_folder = trees_folder, |
| filename = file_of_cherries, |
| dset_prefix = cherries_dset_prefix, |
| max_len = max_len) |
| |
