Added valid_sample checking in database. Fixed db_save_results to use sample...

Added valid_sample checking in database.  Fixed db_save_results to use sample ids.  Adjusted memory request usage for spp & db_save.

chipathlon/db.py

@@ -39,8 +39,9 @@ class MongoDB(object):
         # Make sure output_file exists
         if os.path.isfile(output_file):
             # Make sure that all control_ids & experiment_ids are valid
-            valid_controls = [self.is_valid_experiment(cid) for cid in control_ids]
-            valid_experiments = [self.is_valid_experiment(eid) for eid in experiment_ids]
+            # REMEMBER, these are ids for control & experiment SAMPLES
+            valid_controls = [self.is_valid_sample(cid) for cid in control_ids]
+            valid_experiments = [self.is_valid_sample(eid) for eid in experiment_ids]
             if all(valid_controls) and all(valid_experiments):
                 # First, we load the output file into gfs
                 with open(output_file, "r") as rh:
@@ -61,9 +62,7 @@ class MongoDB(object):
             else:
                 msg = "Not all input ids are valid.  The following are invalid: "
                 for id_list, valid_list in zip([control_ids, experiment_ids], [valid_controls, valid_experiments]):
-                    for i, valid in enumerate(valid_list):
-                        if not valid:
-                            msg += id_list[i] + ", "
+                    msg += ",".join([id_list[i] for i, valid in enumerate(valid_list) if not valid])
         else:
             msg = "Specified output_file %s does not exist." % (output_file,)
         return (False, msg, None)
@@ -77,6 +76,7 @@ class MongoDB(object):
             # chr, start, end, name, score, strand
             # Load data using a list comprehension over lines,
             # then insert with insert_many()
+            print "loading bed_data..."
             with open(bed_file, "r") as rh:
                 bed_data = [
                     {
@@ -92,6 +92,7 @@ class MongoDB(object):
                     for line_info in (line.split(),)
                 ]
                 try:
+                    print "bed data loaded, inserting."
                     self.db.bed.insert_many(bed_data)
                     return (True, "Bed file successfully inserted.", result_id)
                 except pymongo.errors.OperationFailure as e:
@@ -132,6 +133,17 @@ class MongoDB(object):
                     msg = "Error inserting peak_file %s: %s" % (peak_file, e)
         return (valid, msg, None)
 
+    def is_valid_sample(self, sample_accession):
+        try:
+            cursor = self.db.samples.find({
+                "accession": sample_accession
+            })
+            if cursor.count() == 1:
+                return True
+        except pymongo.errors.OperationFailure as e:
+            print "Error with sample_accession %s: %s" % (sample_accession, e)
+        return False
+
     def is_valid_experiment(self, experiment_id):
         try:
             cursor = self.db.experiments.find({

chipathlon/jobs/params/db_save_result.yaml

@@ -40,5 +40,5 @@ db_save_result:
         has_value: true
         default: $inputs.4
   walltime: 2000
-  memory: 2000
+  memory: 16000
   cores: 1

chipathlon/jobs/params/r_spp_nodups.yaml

@@ -65,5 +65,5 @@ r_spp_nodups:
         has_value: true
         default: 8
   walltime: 2000
-  memory: 8000
+  memory: 16000
   cores: 8

chipathlon/jobs/scripts/db_save_result.py

+#!/usr/bin/env python
 import chipathlon.db
 import argparse
 import yaml
@@ -15,7 +16,13 @@ if os.path.isfile(args.file) and os.path.isfile(args.meta):
     mdb = chipathlon.db.MongoDB(args.host, args.username, args.password)
     with open(args.meta, "r") as rh:
         meta = yaml.load(rh)
+    valid = False
+    msg = ""
     if meta["result_type"] == "bed":
-        mdb.save_bed(args.file, meta["control_ids"], meta["experiment_ids"], meta)
+        valid, msg, data = mdb.save_bed(args.file, meta["control_ids"], meta["experiment_ids"], meta)
     elif meta["result_type"] == "peak":
-        mdb.save_peak(args.file, meta["control_ids"], meta["experiment_ids"], meta)
+        valid, msg, data = mdb.save_peak(args.file, meta["control_ids"], meta["experiment_ids"], meta)
+    print msg 
+        
+else:
+    print "Either input file %s or meta file %s does not exist." % (args.file, args.meta)