Mercurial > repos > bgruening > sucos_clustering
comparison sucos_max.py @ 4:791c86130585 draft
"planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/sucos commit c35334ca80c87a5078da1a6df85b34e23b80d837"
| author | bgruening |
|---|---|
| date | Wed, 15 Apr 2020 09:26:30 -0400 |
| parents | bd12f4b4c3a8 |
| children | b8725fec8c7b |
comparison
equal
deleted
inserted
replaced
| 3:bd12f4b4c3a8 | 4:791c86130585 |
|---|---|
| 37 import sucos, utils | 37 import sucos, utils |
| 38 import argparse, gzip, os | 38 import argparse, gzip, os |
| 39 from rdkit import Chem | 39 from rdkit import Chem |
| 40 | 40 |
| 41 | 41 |
| 42 def process(inputfilename, clusterfilenames, outputfilename): | 42 def process(inputfilename, clusterfilenames, outputfilename, filter_value, filter_field): |
| 43 all_clusters = {} | 43 all_clusters = {} |
| 44 for filename in clusterfilenames: | 44 for filename in clusterfilenames: |
| 45 cluster = [] | 45 cluster = [] |
| 46 cluster_file = utils.open_file_for_reading(filename) | 46 cluster_file = utils.open_file_for_reading(filename) |
| 47 suppl = Chem.ForwardSDMolSupplier(cluster_file) | 47 suppl = Chem.ForwardSDMolSupplier(cluster_file) |
| 78 except: | 78 except: |
| 79 utils.log("WARNING: failed to generate features for molecule", mol_num, "in input") | 79 utils.log("WARNING: failed to generate features for molecule", mol_num, "in input") |
| 80 continue | 80 continue |
| 81 scores_max = [0, 0, 0] | 81 scores_max = [0, 0, 0] |
| 82 scores_cum = [0, 0, 0] | 82 scores_cum = [0, 0, 0] |
| 83 cluster_name = None | |
| 83 for clusterfilename in all_clusters: | 84 for clusterfilename in all_clusters: |
| 84 cluster = all_clusters[clusterfilename] | 85 cluster = all_clusters[clusterfilename] |
| 85 index = 0 | 86 index = 0 |
| 86 for entry in cluster: | 87 for entry in cluster: |
| 87 hit = entry[0] | 88 hit = entry[0] |
| 102 scores_cum[0] += sucos_score | 103 scores_cum[0] += sucos_score |
| 103 scores_cum[1] += fm_score | 104 scores_cum[1] += fm_score |
| 104 scores_cum[2] += vol_score | 105 scores_cum[2] += vol_score |
| 105 | 106 |
| 106 | 107 |
| 107 cluster_file_name_only = cluster_name.split(os.sep)[-1] | |
| 108 | |
| 109 # utils.log("Max SuCOS:", scores[0], "FM:", scores[1], "P:", scores[2],"File:", cluster_file_name_only, "Index:", cluster_index) | 108 # utils.log("Max SuCOS:", scores[0], "FM:", scores[1], "P:", scores[2],"File:", cluster_file_name_only, "Index:", cluster_index) |
| 110 mol.SetDoubleProp("Max_SuCOS_Score", scores_max[0] if scores_max[0] > 0 else 0) | 109 mol.SetDoubleProp("Max_SuCOS_Score", scores_max[0] if scores_max[0] > 0 else 0) |
| 111 mol.SetDoubleProp("Max_SuCOS_FeatureMap_Score", scores_max[1] if scores_max[1] > 0 else 0) | 110 mol.SetDoubleProp("Max_SuCOS_FeatureMap_Score", scores_max[1] if scores_max[1] > 0 else 0) |
| 112 mol.SetDoubleProp("Max_SuCOS_Protrude_Score", scores_max[2] if scores_max[2] > 0 else 0) | 111 mol.SetDoubleProp("Max_SuCOS_Protrude_Score", scores_max[2] if scores_max[2] > 0 else 0) |
| 113 mol.SetProp("Max_SuCOS_Cluster", cluster_file_name_only) | 112 |
| 114 mol.SetIntProp("Max_SuCOS_Index", cluster_index) | 113 if cluster_name: |
| 114 cluster_file_name_only = cluster_name.split(os.sep)[-1] | |
| 115 mol.SetProp("Max_SuCOS_Cluster", cluster_file_name_only) | |
| 116 mol.SetIntProp("Max_SuCOS_Index", cluster_index) | |
| 115 | 117 |
| 116 # utils.log("Cum SuCOS:", scores[0], "FM:", scores[1], "P:", scores[2]) | 118 # utils.log("Cum SuCOS:", scores[0], "FM:", scores[1], "P:", scores[2]) |
| 117 mol.SetDoubleProp("Cum_SuCOS_Score", scores_cum[0] if scores_cum[0] > 0 else 0) | 119 mol.SetDoubleProp("Cum_SuCOS_Score", scores_cum[0] if scores_cum[0] > 0 else 0) |
| 118 mol.SetDoubleProp("Cum_SuCOS_FeatureMap_Score", scores_cum[1] if scores_cum[1] > 0 else 0) | 120 mol.SetDoubleProp("Cum_SuCOS_FeatureMap_Score", scores_cum[1] if scores_cum[1] > 0 else 0) |
| 119 mol.SetDoubleProp("Cum_SuCOS_Protrude_Score", scores_cum[2] if scores_cum[2] > 0 else 0) | 121 mol.SetDoubleProp("Cum_SuCOS_Protrude_Score", scores_cum[2] if scores_cum[2] > 0 else 0) |
| 120 | 122 |
| 121 writer.write(mol) | 123 if filter_value and filter_field: |
| 124 if mol.HasProp(filter_field): | |
| 125 val = mol.GetDoubleProp(filter_field) | |
| 126 if val > filter_value: | |
| 127 writer.write(mol) | |
| 128 else: | |
| 129 writer.write(mol) | |
| 122 | 130 |
| 123 input_file.close() | 131 input_file.close() |
| 124 writer.flush() | 132 writer.flush() |
| 125 writer.close() | 133 writer.close() |
| 126 output_file.close() | 134 output_file.close() |
| 133 def main(): | 141 def main(): |
| 134 parser = argparse.ArgumentParser(description='Max SuCOS scores with RDKit') | 142 parser = argparse.ArgumentParser(description='Max SuCOS scores with RDKit') |
| 135 parser.add_argument('-i', '--input', help='Input file to score in SDF format. Can be gzipped (*.gz).') | 143 parser.add_argument('-i', '--input', help='Input file to score in SDF format. Can be gzipped (*.gz).') |
| 136 parser.add_argument('-o', '--output', help='Output file in SDF format. Can be gzipped (*.gz).') | 144 parser.add_argument('-o', '--output', help='Output file in SDF format. Can be gzipped (*.gz).') |
| 137 parser.add_argument('clusters', nargs='*', help="One or more SDF files with the clustered hits") | 145 parser.add_argument('clusters', nargs='*', help="One or more SDF files with the clustered hits") |
| 146 parser.add_argument('--filter-value', type=float, help='Filter out values with scores less than this.') | |
| 147 parser.add_argument('--filter-field', help='Field to use to filter values.') | |
| 138 | 148 |
| 139 args = parser.parse_args() | 149 args = parser.parse_args() |
| 140 utils.log("Max SuCOS Args: ", args) | 150 utils.log("Max SuCOS Args: ", args) |
| 141 | 151 |
| 142 process(args.input, args.clusters, args.output) | 152 process(args.input, args.clusters, args.output, args.filter_value, args.filter_field) |
| 143 | 153 |
| 144 | 154 |
| 145 if __name__ == "__main__": | 155 if __name__ == "__main__": |
| 146 main() | 156 main() |