get_scores_singleconf.py

from casp_rna_em.run_metric_programs import run_phenix_clashscore,run_phenix_rna_validate,run_phenix_cc,run_atomic_inclusion,run_tempy,run_Qscore
from tqdm import tqdm
from glob import glob
import pandas as pd

chimera_location = '/home/groups/rhiju/rkretsch/chimera_1.15'
mapq_script = '/home/groups/rhiju/rkretsch/CASP15_RNA_EM/mapq/mapq/mapq_cmd.py'
phenix_location = '/home/groups/rhiju/rkretsch/phenix/phenix-1.18.2-3874/build/bin/'
chimerax_location = '/home/groups/rhiju/rkretsch/chimerax-1.3/bin/'

l1=['auto-drrafter_erraser','auto-drrafter_only']
l2=['SARS-CoV-1'] 
maps=['SARS_CoV_1-7p05-cryosparc_P18_J109_map_sharp.mrc']
res=[7.05]
thresh=[0.1]
l3=['eternafold','icSHAPE'] 

for a1 in l1:
    for a2,emmap,resolution,threshold in zip(l32,maps,res,thresh):
        for a3 in l3:
            if a1 == 'auto-drrafter_erraser':
                pdbs = glob(f'../../models/{a1}/{a2}/{a3}/*_all_models_all_fits_FINAL_R?.out.??.pdbFINISHED_1.pdb') + glob(f'../../models/{a1}/{a2}/{a3}/*_all_models_all_fits_FINAL_R?.out.?.pdbFINISHED_1.pdb')
            else:
                pdbs = glob(f'../../models/{a1}/{a2}/{a3}/*_all_models_all_fits_FINAL_R?.out.??.pdb') + glob(f'../../models/{a1}/{a2}/{a3}/*_all_models_all_fits_FINAL_R?.out.?.pdb')
            out_file_prefix = f'{a1}_{a2}_{a3}_{a4}'

            single_scores = {}
            per_residue_scores = {}
            per_threshold_scores = {}

            for pdb in tqdm(pdbs):
                print(pdb)
                score_dict = {}
                score_dict.update(run_Qscore(pdb, emmap, mapq_script, chimera_location, resolution))
                score_dict.update(run_phenix_clashscore(
                    pdb, phenix_location=phenix_location))
                score_dict.update(run_phenix_rna_validate(
                    pdb, phenix_location=phenix_location))
                score_dict.update(run_phenix_cc(
                            pdb, emmap, resolution=resolution,
                            phenix_location=phenix_location))
                score_dict.update(run_atomic_inclusion(
                    pdb, emmap, threshold, chimerax_location))
                score_dict.update(run_tempy(pdb, emmap, resolution))

                single_score = {
                    key: score_dict[key] for key in score_dict if "per_residue" not in key and "per_threshold" not in key}
                per_residue_score = {key: score_dict[key]
                                     for key in score_dict if "per_residue" in key}
                per_threshold_score = {key: score_dict[key]
                                       for key in score_dict if "per_threshold" in key}

                model_name = pdb.rsplit(".", 1)[0].rsplit('/', 1)[-1]
                single_scores[model_name] = single_score
                per_residue_scores[model_name] = per_residue_score
                per_threshold_scores[model_name] = per_threshold_score

            df = pd.DataFrame(single_scores).transpose()
            df.index.name = 'pdb'
            df = df.reset_index()
            df['emmap'] = emmap.rsplit('/', 1)[-1]
            df.to_csv(f"{out_file_prefix}_scores.csv", index=False)

            per_res = pd.DataFrame(per_residue_scores).transpose()
            per_res.index.name = 'model'
            if len(per_res.columns) > 0:
                per_res = per_res.explode(list(per_res.columns))
            per_res.to_csv(f"{out_file_prefix}_per_residue.csv")

            per_thr = pd.DataFrame(per_threshold_scores).transpose()
            per_thr.index.name = 'model'
            if len(per_thr.columns) > 0:
                per_thr = per_thr.explode(list(per_thr.columns))
            per_thr.to_csv(f"{out_file_prefix}_per_threshold.csv")