Python 3 fixes (#26)

* Read gzipped or uncompressed files with try/except.
* Use explicit int() for divisions that used to produce integers in python2 -- e.g. in binary search algorithms where the result is used for indexing.
* Exit the top-level script with an error code if any component script fails.

scripts/Assemblytics

@@ -86,7 +86,7 @@ then
     then
         echo "BETWEEN,DONE,Step 2: Assemblytics_between_alignments.pl completed successfully. Now finding variants within alignments." >> $LOG_FILE
         >&2 echo "3. Finding variants within alignments"
-        $SCRIPT_PATH/Assemblytics_within_alignment.py --delta $OUTPUT_PREFIX.Assemblytics.unique_length_filtered_l$UNIQUE_LENGTH.delta.gz --min $MINIMUM_SIZE > $OUTPUT_PREFIX.variants_within_alignments.bed
+        $SCRIPT_PATH/Assemblytics_within_alignment.py --delta $OUTPUT_PREFIX.Assemblytics.unique_length_filtered_l$UNIQUE_LENGTH.delta.gz --min $MINIMUM_SIZE --output $OUTPUT_PREFIX.variants_within_alignments.bed
 
         if [ -e $OUTPUT_PREFIX.variants_within_alignments.bed ];
         then
@@ -123,19 +123,21 @@ then
                     echo "SUMMARY,DONE,Step 5: Assemblytics_summary.py completed successfully" >> $LOG_FILE
                 else
                     echo "SUMMARY,FAIL,Step 5: Assemblytics_summary.py failed" >> $LOG_FILE
+                    exit 1
                 fi
-
             else
                 echo "COMBINE,FAIL,Step 4: combining variants failed" >> $LOG_FILE
+                exit 1
             fi
         else
-
             echo "WITHIN,FAIL,Step 3: Assemblytics_within_alignment.py failed: Possible problem before this step or with Python on server." >> $LOG_FILE
+            exit 1
         fi
     else
         echo "BETWEEN,FAIL,Step 2: Assemblytics_between_alignments.pl failed: Possible problem with Perl or show-coords on server." >> $LOG_FILE
+        exit 1
     fi
 else
     echo "UNIQFILTER,FAIL,Step 1: Assemblytics_uniq_anchor.py failed: Possible problem with Python or Python packages on server." >> $LOG_FILE
+    exit 1
 fi
-

scripts/Assemblytics_uniq_anchor.py

@@ -25,21 +25,21 @@ def run(args):
     if unique_length == 10000:
         print("Use --unique-length X to set the unique anchor length requirement. Default is 10000, such that each alignment must have at least 10000 bp from the query that are not included in any other alignments.")
 
-
-    print("header:")
-
-    f = open(filename)
-    header1 = f.readline()
-    # the first two bytes show whether a file is gzipped
-    if header1[0:2]=="\x1f\x8b":
-        f.close()
-        f = gzip.open(filename)
-        print(f.readline().strip())
-    else:
-        print(header1.strip())
-
+    try:
+        f = gzip.open(filename, 'rt')
+        header1 = f.readline().strip()
+        print("Detected gzipped delta file. Reading...")
+    except:
+        f = open(filename, 'r')
+        header1 = f.readline().strip()
+        print("Detected uncompressed delta file. Reading...")
+
     # Ignore the first two lines for now
+    print("\n")
+    print("Header (2 lines):")
+    print(header1)
     print(f.readline().strip())
+    print("\n")
 
     linecounter = 0
 
@@ -56,7 +56,6 @@ def run(args):
 
     for line in f:
         if line[0]==">":
-
             fields = line.strip().split()
             current_query_name = fields[1]
             current_header = line.strip()
@@ -84,9 +83,9 @@ def run(args):
     num_queries = len(lines_by_query)
     print("Filtering alignments of %d queries" % (num_queries))
 
-    num_query_step_to_report = num_queries/100
+    num_query_step_to_report = int(num_queries/100)
     if num_queries < 100:
-        num_query_step_to_report = num_queries/10
+        num_query_step_to_report = int(num_queries/10)
     if num_queries < 10:
         num_query_step_to_report = 1
 
@@ -97,22 +96,24 @@ def run(args):
 
         query_counter += 1
         if (query_counter % num_query_step_to_report) == 0:
-            print("Progress: %d%%" % (query_counter*100/num_queries))
+            print("Progress: %d%%" % (int(query_counter*100/num_queries)))
     print("Progress: 100%")
 
     print("Deciding which alignments to keep: %d seconds for %d queries" % (time.time()-before,num_queries))
     before = time.time()
 
-    fout = gzip.open(output_filename + ".Assemblytics.unique_length_filtered_l%d.delta.gz" % (unique_length),'w')
+    fout = gzip.open(output_filename + ".Assemblytics.unique_length_filtered_l%d.delta.gz" % (unique_length),'wt')
 
-    f = open(filename)
-    header1 = f.readline()
-    if header1[0:2]=="\x1f\x8b":
-        f.close()
-        f = gzip.open(filename)
-        header1 = f.readline()
-
-    fout.write(header1) # write the first line that we read already
+    try:
+        f = gzip.open(filename, 'rt')
+        header1 = f.readline().strip()
+        print("Detected gzipped delta file. Reading...")
+    except:
+        f = open(filename, 'r')
+        header1 = f.readline().strip()
+        print("Detected uncompressed delta file. Reading...")
+
+    fout.write(header1)
     fout.write(f.readline())
 
     linecounter = 0
@@ -329,7 +330,7 @@ def binary_search(query, numbers, left, right):
 
     if left >= right:
         return right
-    mid = (right+left)/2
+    mid = int((right+left)/2)
 
 
     if query == numbers[mid]:

scripts/Assemblytics_variant_charts.R

@@ -34,8 +34,6 @@ comma_format<-function(num) {
     formatC(abs(num),format="f",big.mark=",",drop0trailing = TRUE)
 }
 
-
-
 # Prep data for log-scaled plot
 alt <- bed
 

scripts/Assemblytics_within_alignment.py

@@ -12,16 +12,21 @@ def run(args):
     filename = args.delta
     minimum_variant_size = args.minimum_variant_size
 
-    f = open(filename)
-    header1 = f.readline()
-    # check first 2 bytes to see if the file is gzipped
-    if header1[0:2]=="\x1f\x8b":
-        f.close()
-        f = gzip.open(filename)
-        header1 = f.readline()
-
+    try:
+        f = gzip.open(filename, 'rt')
+        header1 = f.readline().strip()
+        print("Detected gzipped delta file. Reading...")
+    except:
+        f = open(filename, 'r')
+        header1 = f.readline().strip()
+        print("Detected uncompressed delta file. Reading...")
+
     # Ignore the first two lines for now
-    f.readline()
+    print("\n")
+    print("Header (2 lines):")
+    print(header1)
+    print(f.readline().strip())
+    print("\n")
 
     linecounter = 0
 
@@ -74,18 +79,20 @@ def run(args):
 
     f.close()
 
+    fout = open(args.output_path, 'w')
     newcounter = 1
     for line in variants:
         if line[4] >= minimum_variant_size:
             line[3] = "Assemblytics_w_%d" % (newcounter)
-            print("\t".join(map(str,line[0:10])) + ":" + str(line[11]) + "-" + str(line[12]) + ":+\t" + line[10])
+            fout.write("\t".join(map(str,line[0:10])) + ":" + str(line[11]) + "-" + str(line[12]) + ":+\t" + line[10] + "\n")
             newcounter += 1
-
+    fout.close()
 
 def main():
     parser=argparse.ArgumentParser(description="Outputs MUMmer coordinates annotated with length of unique sequence for each alignment")
     parser.add_argument("--delta",help="delta file" ,dest="delta", type=str, required=True)
     parser.add_argument("--min",help="Minimum size (bp) of variant to include, default = 50" ,dest="minimum_variant_size",type=int, default=50)
+    parser.add_argument("--output", help="Output file with variants in bed format.", dest="output_path", type=str, required=True)
     parser.set_defaults(func=run)
     args=parser.parse_args()
     args.func(args)