generic splitbc component

workflows/radseq/components/splitbc.py → workflows/components/splitbc.py

@@ -19,54 +19,58 @@ import os
 
 from jflow.iotypes import OutputFileList,OutputFile, InputFile, InputFileList,  Formats
 from jflow.abstraction import MultiMap
+from jflow.component import Component
 
 from weaver.function import ShellFunction
-from ng6.analysis import Analysis
 
-class SplitBC (Analysis):
-    
-    ENZYMES = {
-              'sbfI' : {
-                        'rad'    : 'CCTGCAGG',
-                        'radtag' : 'TGCAGG'
-            }
-    }
-    
-    def get_enzyme (self, name):
-        if not self.ENZYMES.has_key(name) :
-            raise ValueError("The enzyme name " + str(name) + " does not exists. Accepted names are " + str (self.ENZYMES.keys()))
-    
-        return (self.ENZYMES[name]['rad'], self.ENZYMES[name]['radtag'])
+
+class SplitBC (Component):
     
-    def define_parameters(self, fastq_file1, fastq_file2, matrix_indiv_name, barcode_file, enzyme,
-                          mismatches, tag_mismatch, trim = True, forward = True ):
+    def define_parameters(self, fastq_file1, barcode_file, matrix_indiv_name, fastq_file2 = [], 
+                          rad = None, rad_tag = None, mismatches = None , tag_mismatch = None , trim_barcode = False, 
+                          trim_reads2 = False, bol = True, eol = False, partial = None, no_adapt = False):
         """
             @param fastq_file1: list of fastq_files path
+            @param barcode_file: list of barcode file path
+            @param matrix_indiv_name: list of list of individual names (example : [[], [] , []]). The length of this
+                matrix_indiv_name must be the same as the length of fastq_file1 (and fastq_file2 if provided)
             @param fastq_file2: list of fastq_files path
-            @param matrix_indiv_name: list of list of individual names
-            @param barcode_file: list of bascode file path
-            @param enzyme: enzyme name
+            @param rad:
+            @param rad_tag: 
+            ...
         """
-        
         check_len = len(fastq_file1) == len(matrix_indiv_name) == len(barcode_file)
-        
         self.fastq1 = OutputFileList(fastq_file1, Formats.FASTQ)
         self.fastq2 = None
-        if fastq_file2 is not None:
+        
+        if fastq_file2 :
             check_len = len(fastq_file1) == len(matrix_indiv_name) == len(fastq_file1) == len(barcode_file)
             self.fastq2 = OutputFileList(fastq_file2, Formats.FASTQ)
         
         if not check_len :
             raise Exception("length of fastq_file1, fastq_file2, matrix_indiv_name and barcode_file must be the same")
         
+        if trim_barcode and trim_reads2 :
+            raise Exception("you must specify either trim_barcode or trim_reads2, but not both")
+
+        if (bol == eol == True ) or ( bol == eol == False) :
+            raise Exception("one of bol, eol must be specified, but not both")
+        
+        if (rad is not None and rad_tag is None) or (rad is None and rad_tag is not None) :
+            raise Exception("you must specify rad with rad_tag")
+        
         self.barcode_file = OutputFileList(barcode_file)
         self.mismatches = mismatches
-        self.rad, self.rad_tag = self.get_enzyme(enzyme )
         self.tag_mismatch = tag_mismatch
-        self.trim = trim
-        self.forward = forward
-        self.prefix_r1 = os.path.join(self.output_directory , "%_1.fq")
-        self.prefix_r2 = os.path.join(self.output_directory , "%_2.fq")
+        self.trim_barcode = trim_barcode
+        self.trim_reads2 = trim_reads2
+        self.bol = bol
+        self.eol = eol
+        self.partial = partial
+        self.no_adapt = no_adapt
+        self.rad = rad
+        self.rad_tag = rad_tag
+        
         self.matrix_read1 = []
         self.matrix_read2 = []
         self.output_read1 = []
@@ -75,46 +79,54 @@ class SplitBC (Analysis):
         
         for id, inames in enumerate(matrix_indiv_name) :
             outr1 = OutputFileList(self.get_outputs('{basename_woext}_1.fq', inames), Formats.FASTQ)
-            outr2 = OutputFileList(self.get_outputs('{basename_woext}_2.fq', inames), Formats.FASTQ)
             self.matrix_read1.append(outr1)
-            self.matrix_read2.append(outr2)
             self.output_read1 += outr1
-            self.output_read2 += outr2
-            self.stdout.append(OutputFile(os.path.join(self.output_directory , "splitbc" + str(id) + ".stdout")))
-
-    
-    def get_version(self):
-        return "1.1"
- 
- 
-    def define_analysis(self):
-        self.name = "splitbc"
-        self.description = "Demultiplex individual"
-        self.software = "splitbc.pl"
-        self.options =  ' '
+            
+            if self.fastq2 is not None :
+                outr2 = OutputFileList(self.get_outputs('{basename_woext}_2.fq', inames), Formats.FASTQ)
+                self.matrix_read2.append(outr2)
+                self.output_read2 += outr2
 
+            self.stdout.append(OutputFile(os.path.join(self.output_directory , "splitbc" + str(id) + ".stdout")))
 
     def process(self):
-        strand = "--bol" if self.forward else "--eol"
+        prefix_r1 = os.path.join(self.output_directory , "%_1.fq")
+        command = [self.get_exec_path("splitbc.pl"), "--bcfile", "$1", "--prefix-r1", prefix_r1 ]
+        
+        if self.bol :
+            command.append('--bol')
+        elif self.eol :
+            command.append('--eol')
+        
+        if self.mismatches is not None :
+            command.extend(["--mismatches", self.mismatches])
+        
+        if self.tag_mismatch is not None :
+            command.extend(["--TAG_mismatch", self.tag_mismatch])
+        
+        if self.partial is not None :
+            command.extend(["--partial", self.partial])
+        
+        if self.trim_barcode :
+            command.append( "--trim")
+        elif self.trim_reads2 :
+            command.append( "--trim2")
+            
+        if self.no_adapt is not None :
+            command.append("--no_adapt")
+        
+        if self.rad is not None and self.rad_tag is not None :
+            command.extend(["--rad", self.rad, "--radTAG", self.rad_tag])
+        
         if self.fastq2 is not None :
-            command = " ".join([
-                self.get_exec_path("splitbc.pl"), "$1", "$2", "--mismatches", self.mismatches,
-                "--bcfile", "$3", "--rad", self.rad, "--radTAG", self.rad_tag, "--TAG_mismatch", self.tag_mismatch,
-                "--trim" if self.trim else "", "--prefix-r1", self.prefix_r1, "--prefix-r2", self.prefix_r2, strand , ' 2>&1 >> $4 '
-            ])
-    
+            prefix_r2 = os.path.join(self.output_directory , "%_2.fq")
+            command.extend(["--prefix-r2", prefix_r2, "$2", "$3", '2>&1 >> $4' ])
+            command = ' '.join(command)
             splitbc = ShellFunction(command, cmd_format='{EXE} {IN} {OUT}')
-            MultiMap(splitbc, inputs=[self.fastq1, self.fastq2, self.barcode_file], outputs=[self.stdout, self.matrix_read1, self.matrix_read2])
+            MultiMap(splitbc, inputs=[self.barcode_file, self.fastq1, self.fastq2], outputs=[self.stdout, self.matrix_read1, self.matrix_read2])
+
         else :
-            command = " ".join([
-                self.get_exec_path("splitbc.pl"), "$1","--mismatches", self.mismatches,
-                "--bcfile", "$2", "--rad", self.rad, "--radTAG", self.rad_tag, "--TAG_mismatch", self.tag_mismatch,
-                "--trim" if self.trim else "", "--prefix-r1", self.prefix_r1, strand, ' 2>&1 > $3 '
-            ])
-    
+            command.extend([ "$2", '2>&1 >> $3' ])
+            command = ' '.join(command)
             splitbc = ShellFunction(command, cmd_format='{EXE} {IN} {OUT}')
-            MultiMap(splitbc, inputs=[self.fastq1, self.barcode_file], outputs=[self.stdout, self.matrix_read1])
-
-        
-    def post_process(self):
-        print ""
\ No newline at end of file
+            MultiMap(splitbc, inputs=[self.barcode_file, self.fastq1], outputs=[self.stdout, self.matrix_read1])

workflows/radseq/__init__.py

@@ -24,8 +24,23 @@ from ng6.ng6workflow import NG6Workflow
 
 class RADseq (NG6Workflow):
     
+    ENZYMES = {
+              'sbfI' : {
+                        'rad'    : 'CCTGCAGG',
+                        'radtag' : 'TGCAGG'
+            }
+    }
+        
+    def get_enzyme (self, name):
+        if not self.ENZYMES.has_key(name) :
+            raise ValueError("The enzyme name " + str(name) + " does not exists. Accepted names are " + str (self.ENZYMES.keys()))
+    
+        return (self.ENZYMES[name]['rad'], self.ENZYMES[name]['radtag'])
+    
+    
     def process(self):
         
+        rad, rad_tag = self.get_enzyme(self.args['enzyme'])
         # group all individual by pool
         pools = {}
         for p in self.args['pool'] :
@@ -65,10 +80,11 @@ class RADseq (NG6Workflow):
                     ff.write(indiv['indiv_name'] + "\t" + indiv['barcode'] +"\n")
             indiv_names.append(inames)
         
-        splitbc = self.add_component("SplitBC", [ fastq_files_1,fastq_files_2 if fastq_files_2 else None,
-                      indiv_names, barcode_files, self.args['enzyme'], self.args['mismatches'], self.args['tag_mismatch'],
-                      self.args['trim'], self.args['forward']])
+
+        splitbc = self.add_component("SplitBC", [ fastq_files_1,barcode_files, indiv_names, fastq_files_2,
+                      rad, rad_tag, self.args['mismatches'], self.args['tag_mismatch'], 
+                      self.args['trim_barcode'], self.args['trim_reads2']])
       
-        ustacks = self.add_component("Ustacks" , [], {"indiv_dic": indivs_by_name, "read1_files" : splitbc.output_read1 , "max_locus" : 3 } )
+        #ustacks = self.add_component("Ustacks" , [], {"indiv_dic": indivs_by_name, "read1_files" : splitbc.output_read1 , "max_locus" : 3 } )
         #cstacks = self.add_component("Cstacks", [ustacks.alleles, ustacks.snps, ustacks.tags, self.args["catalog_mismatches"]])
         

workflows/radseq/workflow.properties

@@ -94,17 +94,19 @@ mismatches.flag = --mismatches
 mismatches.help = Max. number of mismatches allowed.
 mismatches.default = 1
 
-forward.name = forward
-forward.flag = --forward
-forward.help = Match barcode at the begening of the sequence (5' 3')
-forward.default = True
-forward.type = bool
+trim_barcode.name = trim_barcode
+trim_barcode.flag = --trim-barcode
+trim_barcode.help = Should the barecode be trimmed
+trim_barcode.default = True
+trim_barcode.exclude = trim_reads2
+trim_barcode.type = bool
 
-trim.name = trim
-trim.flag = --trim
-trim.help = Should the barecode be trimmed.
-trim.default = True
-trim.type = bool
+trim_reads2.name = trim_reads2
+trim_reads2.flag = --trim-reads2
+trim_reads2.help = Shoud the read 2 be trimmed to have the same length as the read1
+trim_reads2.default = False
+trim_reads2.exclude = trim_barcode
+trim_reads2.type = bool
 
 catalog_mismatches.name = catalog_mismatches
 catalog_mismatches.flag = --catalog-mismatches