Exchange prediction path to origin

nerpa.py

@@ -15,6 +15,7 @@
 site.addsitedir(nerpa_init.python_modules_dir)
 
 import predictions_preprocessor
+import postprocessing
 import nerpa_utils
 import handle_rban
 import logger
@@ -202,7 +203,7 @@ def copy_prediction_list(args, main_out_dir):
     return new_prediction_path
 
 
-def get_antismash_v3_compatible_input_paths(listing_fpath, list_of_paths, output_dir, log):
+def get_antismash_v3_compatible_input_paths(listing_fpath, list_of_paths, output_dir, log, origin_file):
     '''
     Parses all antiSMASH-related options,
     detects all relevant output dirs (either with .json [aS v.5] or with ./txt/ & ./nrpspks_predictions_txt [aS v.3],
@@ -270,7 +271,12 @@ def _is_antiSMASHv5_path(path):
         converted_antiSMASH_v5_paths = convert_antiSMASH_v5(antiSMASHv5_paths +
                                                             ['-o', converted_antiSMASH_v5_outputs_dir, '-m', 'hybrid'])
         antiSMASHv3_paths += converted_antiSMASH_v5_paths
+        for (covant5, origin5) in zip(converted_antiSMASH_v5_paths, antiSMASHv5_paths):
+            origin_file[covant5] = origin_file[origin5] if origin5 in origin_file else origin5
         log.info("\n======= Done with Preprocessing antiSMASH v5 inputs")
+    for path3 in antiSMASHv3_paths:
+        if path3 not in origin_file:
+            origin_file[path3] = path3
 
     return antiSMASHv3_paths
 
@@ -281,6 +287,8 @@ def run(args, log):
     log.set_up_file_handler(output_dir)
     log.start()
 
+    origin_file = {}
+
     if args.predictions is not None:
         path_predictions = copy_prediction_list(args, output_dir)
     else:
@@ -301,10 +309,11 @@ def run(args, log):
                        '--cpus', str(args.threads), args.seqs]
             nerpa_utils.sys_call(command, log, cwd=output_dir)
             antismash_out_dirs.append(cur_antismash_out)
+            origin_file[cur_antismash_out] = args.seqs
 
         path_predictions = predictions_preprocessor.create_predictions_by_antiSAMSHout(get_antismash_v3_compatible_input_paths(
                 listing_fpath=args.antismash_out, list_of_paths=antismash_out_dirs,
-                output_dir=output_dir, log=log), output_dir, log)
+                output_dir=output_dir, log=log, origin_file=origin_file), output_dir, log, origin_file)
 
     input_configs_dir = args.configs_dir if args.configs_dir else nerpa_init.configs_dir
     current_configs_dir = os.path.join(output_dir, "configs")
@@ -345,6 +354,9 @@ def run(args, log):
     log.info("RESULTS:")
     log.info("Main report is saved to " + os.path.join(output_dir, 'report.csv'), indent=1)
     log.info("Detailed reports are saved to " + output_dir, indent=1)
+
+    postprocessing.postproccesing(output_dir, origin_file)
+
     log.finish()
 
 

src/nerpa_pipeline/postprocessing.py

@@ -0,0 +1,49 @@
+#!/usr/bin/env python3
+import csv
+import os
+
+def originPathInReport(output_dir, origin_file):
+    lines = []
+    with open(os.path.join(output_dir, "report.csv"), "r") as f:
+        csv_reader = csv.reader(f)
+        for row in csv_reader:
+            lines.append(row)
+
+    pos_id = lines[0].index("PredictionFileName")
+    lines[0] = lines[0][:pos_id + 1] + ["ContigId", "PartId"] + lines[0][pos_id + 1:]
+    for i in range(1, len(lines)):
+        curPath = lines[i][pos_id]
+        orgPath = origin_file[curPath]
+        ctgid = curPath.split('/')[-1].split('_')[-4]
+        partid = curPath.split('/')[-1].split('_')[-1]
+
+        lines[i] = lines[i][:pos_id] + [orgPath, ctgid, partid] + lines[i][pos_id + 1:]
+
+    with open(os.path.join(output_dir, "report.csv"), "w") as fw:
+        writer = csv.writer(fw, delimiter=',')
+        for line in lines:
+            writer.writerow(line)
+
+
+def originPathInDetailReport(output_dir, origin_file):
+    for filename in os.listdir(os.path.join(output_dir, "details")):
+        if filename.endswith('.match'):
+            lines = []
+            with open(os.path.join(output_dir, "details", filename)) as f:
+                for line in f:
+                    if line[:-1] in origin_file:
+                        orgPath = origin_file[line[:-1]]
+                        ctgid = line.split('/')[-1].split('_')[-4]
+                        partid = line.split('/')[-1].split('_')[-1]
+                        lines.append(orgPath + "\t" + ctgid + "\t" + partid)
+                    else:
+                        lines.append(line)
+
+            with open(os.path.join(output_dir, "details", filename), "w") as fw:
+                for line in lines:
+                    fw.write(line)
+
+
+def postproccesing(output_dir, origin_file):
+    originPathInReport(output_dir, origin_file)
+    originPathInDetailReport(output_dir, origin_file)

src/nerpa_pipeline/predictions_preprocessor.py

@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 import sys
 import os
 import shutil
@@ -40,7 +40,7 @@ def gen_prediction_dict(orf_part, input_file_name, dirname):
 
 
 def gen_prediction_for_one_orfs_part(orf_part, input_file_name, output_prefix,
-                                     current_part, predictions_info_list, base_antismashout_name):
+                                     current_part, predictions_info_list, base_antismashout_name, origin_file):
     prediction_dict = gen_prediction_dict(orf_part, input_file_name, base_antismashout_name)
 
     output_str = ""
@@ -53,17 +53,18 @@ def gen_prediction_for_one_orfs_part(orf_part, input_file_name, output_prefix,
         with open(output_file, 'w') as wf:
             wf.write(output_str)
         current_part += 1
+        origin_file[output_file] = origin_file['/'.join(input_file_name.split('/')[:-2])]
         predictions_info_list.append(output_file)
 
     return current_part
 
 def gen_predictions(bgc_orfs_parts, input_file_name, output_prefix, current_part,
-                    predictions_info_list, base_antismashout_name):
+                    predictions_info_list, base_antismashout_name, origin_file):
     for orf_part in bgc_orfs_parts:
-        current_part = gen_prediction_for_one_orfs_part(orf_part, input_file_name, output_prefix, current_part, predictions_info_list, base_antismashout_name)
+        current_part = gen_prediction_for_one_orfs_part(orf_part, input_file_name, output_prefix, current_part, predictions_info_list, base_antismashout_name, origin_file)
     return current_part
 
-def create_predictions_by_antiSAMSHout(antismashouts, outdir, log):
+def create_predictions_by_antiSAMSHout(antismashouts, outdir, log, origin_file):
     log.info("Start create predictions by antiSMASH")
 
     dir_for_predictions = os.path.join(outdir, "predictions")
@@ -111,7 +112,7 @@ def create_predictions_by_antiSAMSHout(antismashouts, outdir, log):
                     #shutil.copyfile(os.path.join(nrpspred_dir, filename), os.path.join(dir_for_predictions, base_antismashout_name + "_" + base_pred_name))
                     gen_predictions(parts, os.path.join(nrpspred_dir, filename),
                                     os.path.join(dir_for_predictions, base_antismashout_name + "_" + base_pred_name)[:-4],
-                                    0, predictions_info_list, dirname)
+                                    0, predictions_info_list, dirname, origin_file)
 
     f = open(predictions_info_file, 'w')
     for line in predictions_info_list: