add function for pooled ONT samples

scripts/ont_stats.py

@@ -5,22 +5,46 @@
 import os
 from collections import OrderedDict
 
-# TODO check for multiple run
-def get_read_length_and_summary(file_path):
-    summary_metrix = pd.read_csv(file_path, delimiter = "\t")
-    read_length = summary_metrix[summary_metrix["passes_filtering"]]["sequence_length_template"].tolist()
-    read_length.sort(reverse = True)
-    median = statistics.median(read_length)
-    N50_value = sum(read_length) / 2
-    total = 0
-    for item in read_length:
-        total += item
-        if total >= N50_value:
-            N50 = item
-            break
-
+# TODO get barcode info from lims
+# check if the run is pooled
+def if_pooled(sequencing_summary_df):
+    pooled = False
+    if "barcode_kit" in sequencing_summary_df.columns:
+        pooled = True
+    return pooled
+
+# get stats metric if the run is not pooled
+def get_read_length_and_summary(sequencing_summary_df):
+    read_length = sequencing_summary_df[sequencing_summary_df["passes_filtering"]]["sequence_length_template"].tolist()
+    if len(read_length) != 0:
+        read_length.sort(reverse = True)
+        median = statistics.median(read_length)
+        N50_value = sum(read_length) / 2
+        total = 0
+        for item in read_length:
+            total += item
+            if total >= N50_value:
+                N50 = item
+                break
+    else:
+        median = 0
+        N50_value = 0
+        N50 = 0
     return(len(read_length), N50_value * 2 / 1000000000, N50, median)
 
+# get stats metric if the run is pooled
+def get_read_length_and_summary_pooled(sequencing_summary_df, sample_name):
+    sample_dict = {}
+    samples = sequencing_summary_df["barcode_arrangement"].unique()
+    for sample in samples:
+        sample_df = sequencing_summary_df.loc[sequencing_summary_df['barcode_arrangement'] == sample]
+        sample_sub = sample_name + "_" + sample
+        stats = get_read_length_and_summary(sample_df)
+        # only record barcodes with more than 10000 reads
+        if stats[0] > 10000:
+            sample_dict[sample_sub] = get_read_length_and_summary(sample_df)
+    return sample_dict
+
 def write_to_csv(sample_dict):
     file_name = "summary.csv"
     with open(file_name,'w') as file:
@@ -35,12 +59,18 @@ def write_to_csv(sample_dict):
     project_directory = sys.argv[1]
     os.chdir(project_directory)
     sample_list = next(os.walk("."))[1]
-    sample_dict = OrderedDict()
+    sample_dict = {}
     sample_list.sort()
     for sample in sample_list:
         destination = project_directory + "/" + sample
         file = glob.glob(destination + "/*/sequencing_summary_*")
         if len(file) != 0:
-            sample_dict[sample] = get_read_length_and_summary(file[0])
-
+            summary_metrix = pd.read_csv(file[0], delimiter = "\t")
+            pooled = if_pooled(summary_metrix)
+            if pooled:
+                sample_dict_sub = get_read_length_and_summary_pooled(summary_metrix, sample)
+                sample_dict.update(sample_dict_sub)
+            else:
+                sample_dict[sample] = get_read_length_and_summary(summary_metrix)
+
     write_to_csv(sample_dict)