Update files for ONT demultiplex

workflows/ont_qc/__init__.py

@@ -71,12 +71,12 @@ class OntQualityCheck (NG6Workflow):
         
         # add logs
         if len(self.log_files) > 0 :
-            add_log = self.add_component("BasicAnalysis", [self.log_files,"Log Files","Log files generated during primary analysis","-","-","-","gz", "","log.gz"])
+            add_log = self.add_component("BasicAnalysis", [self.log_files,"Log Files","Log files generated during primary analysis","-","-","-","none", "","log.gz"])
         
         addrawfiles = self.add_component("AddRawFiles", [self.runobj, self.get_all_reads(), self.compression])
         ontstat = self.add_component("Run_stats", [self.summary_file, sample_names[0]])
-        if (self.barcoded != None) or (self.barcoded != "no") :
-            demultiplexont = self.add_component("Demultiplex_ONT", [self.get_all_reads() , self.barcoded])
+        if (self.barcoded == "yes"):
+            demultiplexont = self.add_component("Demultiplex_ONT", [self.get_all_reads()])
         if self.trimming == "yes":
             trim_porechop = self.add_component("Trim_porechop", [self.get_all_reads() , "discard_middle"])
         if self.fast5dir != None:

workflows/ont_qc/components/demultiplexont.py

@@ -31,11 +31,11 @@ class Demultiplex_ONT (Analysis):
     def __init__(self, args={}, id=None, function= "process"):
         Analysis.__init__(self, args, id, function)
         
-    def define_parameters(self, fastq_files, barcode_file, archivename="DemultiplexONT_archive.tar"):
+    def define_parameters(self, fastq_files, archivename="DemultiplexONT_archive.tar"):
         self.add_input_file_list( "fastq_files", "fastq_files", default=fastq_files, required=True, file_format = 'fastq')
-        self.add_parameter("barcode_file", "Name of the barcode file", default=barcode_file, required=False , file_format = 'str')
+        #self.add_parameter("barcode_file", "Name of the barcode file", default=barcode_file, required=False , file_format = 'str')
         self.add_parameter("archive_name", "Name of the archive", default=archivename, type='str')
-        self.add_parameter( "run_name", "The name of the run (from total fastq file)", pattern='{basename_woext}', items=self.fastq_files, file_format = "fastq")
+        #self.add_parameter( "run_name", "The name of the run (from total fastq file)", pattern='{basename_woext}', items=self.fastq_files, file_format = "fastq")
         
     def define_analysis(self):
         self.name = "DemultiplexONT"
@@ -53,7 +53,7 @@ class Demultiplex_ONT (Analysis):
         list_stats= []
         with open(stat_file, "r") as f_stat :
             for line in f_stat.readlines():
-                list_stat.append(line.split())
+                list_stats.append(line.split())
                 
         # Registering file's header into a list
         header = list_stats.pop(0)
@@ -74,47 +74,26 @@ class Demultiplex_ONT (Analysis):
     def post_process(self):
         logging.getLogger("jflow").debug("Begin DemultiplexONT.post_process! ont_qc")
         # Create dictionary : key = file name or prefix, value = files path
-        results_files = []
+        stats_dico = self.__parse_stat_file(os.path.join(self.output_directory, "DemultiplexONT.output"))
+        
         # add header of stats
-        group = "statsporechop"
-        self._add_result_element("metrics", "headers", ','.join(["read_trim_start", "read_total_start", "bp_removed_start", "read_trim_end", "read_total_end", "bp_removed_end"]), group)
-
-        print(os.listdir(self.output_directory))
-        for file in os.listdir(self.output_directory):
-            full_file_path = os.path.join(self.output_directory, file)
-            logging.getLogger("jflow").debug("Trimporechop.post_process : full_file_path "+full_file_path)
-            if file.endswith(".fastq"):
-                logging.getLogger("jflow").debug("Trimporechop.post_process match .fastq : full_file_path "+full_file_path)
-                results_files.append(full_file_path)
-            elif file.endswith(".stdout"):
-                logging.getLogger("jflow").debug("Trimporechop.post_process match .stdout: full_file_path "+full_file_path)
-                results_files.append(full_file_path)
-
-                filename = os.path.basename(file).split(".stdout")[0]
-                resultlist = self.__parse_stat_file(full_file_path)
-                read_trim_start = resultlist[0]
-                read_total_start = resultlist[1]
-                bp_removed_start = resultlist[2]
-                read_trim_end = resultlist[3]
-                read_total_end = resultlist[4]
-                bp_removed_end = resultlist[5]
-
-                #add stats for each fastq file
-                self._add_result_element("ont_sample", "read_trim_start", read_trim_start,filename)
-                self._add_result_element("ont_sample", "read_total_start", read_total_start,filename)
-                self._add_result_element("ont_sample", "bp_removed_start", bp_removed_start,filename)
-                self._add_result_element("ont_sample", "read_trim_end", read_trim_end,filename)
-                self._add_result_element("ont_sample", "read_total_end", read_total_end,filename)
-                self._add_result_element("ont_sample", "bp_removed_end", bp_removed_end,filename)
-                
-        #Finaly create and add the archive to the analysis
-        #self._create_and_archive(results_files,self.archive_name)
-        self._archive_files(results_files, "gz")
+        stats_names = ["format",'num_seqs','sum_len','avg_len','min_len','max_len',"N50"]
+        #'Q2', , , 'N50', , , 'format', 'Q1', 'Q3', 'type', 'sum_gap', 'Q30(%)', , 'Q20(%)' 
+        self._add_result_element("metrics", "headers", ','.join(stats_names),"stats_names")
+        
+        # Add stats metrics
+        for fastq in stats_dico:
+            if re.search(".fastq.gz",fastq):
+                fastq_name = os.path.basename(fastq).replace(".fastq.gz","").split('_')[-1]
+            else :
+                fastq_name = os.path.splitext(os.path.basename(fastq))[0].split('_')[-1]
+            for stat in stats_dico[fastq]:
+                self._add_result_element("stats_metrics", stat, stats_dico[fastq][stat],fastq_name)
 
         logging.getLogger("jflow").debug("End DemultiplexONT.post_process! ")
     
     def get_version(self):
-        shell_script = self.get_exec_path("seqkit") + " --version"
+        shell_script = self.get_exec_path("seqkit") + " version | head -n1"
         logging.getLogger("jflow").debug("DemultiplexONT.get_version ! shell_script " + str(shell_script))
         cmd = ["sh","-c",shell_script]
         p = Popen(cmd, stdout=PIPE, stderr=PIPE)
@@ -125,15 +104,23 @@ class Demultiplex_ONT (Analysis):
     def process(self):
         logging.getLogger("jflow").debug("Begin DemultiplexONT.process! ont_qc")
         
-
+        iter = 1
+        str_input = ""
+        str_output = ""
+        for fastq in self.fastq_files:
+            str_input = str_input + " $" + str(iter)
+            iter = iter + 1
+        str_output = " $"+ str(iter)
+            
         # Create cmd
-        self.add_shell_execution(self.get_exec_path("seqkit") +" " + self.options + " > ${" + str() + "}",
+        self.add_shell_execution(self.get_exec_path("seqkit") +" stats --all " + str_input + ">" + str_output,
             cmd_format='{EXE} {IN} {OUT}' ,
             map=False,
-            inputs = self.fastq_files)
+            inputs = [self.fastq_files],
+            outputs = os.path.join(self.output_directory, "DemultiplexONT.output"))
 
         #archive = self.output_directory + '/' + self.archive_name + '.tar.gz'
         #self.add_shell_execution('tar -czf $1 ' + self.output_directory + '/' + '*_trim.fastq ', cmd_format='{EXE} {OUT}', map=False, outputs = archive)
 
-        logging.getLogger("jflow").debug("End Trimporechop.process! ")
+        logging.getLogger("jflow").debug("End Seqkit.process! ")