Switch to version 3.0.1b20

version 3.0.1b19

python/obitools3/apps/optiongroups/__init__.py

@@ -63,6 +63,12 @@ def __addImportInputOption(optionManager):
                      const=b'unite',
                      help="Input file is in UNITE fasta format. If NCBI taxonomy provided with --taxonomy, taxid and scientific name will be added for each sequence.")
 
+    group.add_argument('--sintax-input',
+                     action="store_const", dest="obi:inputformat",
+                     default=None,
+                     const=b'sintax',
+                     help="Input file is in SINTAX fasta format. If NCBI taxonomy provided with --taxonomy, taxid and scientific name will be added for each sequence.")
+
     group.add_argument('--embl-input',
                      action="store_const", dest="obi:inputformat",
                      default=None,

python/obitools3/commands/alignpairedend.pyx

@@ -233,6 +233,10 @@ def run(config):
         PyErr_CheckSignals()
 
         consensus = o_view[i]
+        
+        if two_views:
+            consensus[b"R1_parent"] = forward[i].id
+            consensus[b"R2_parent"] = reverse[i].id
                
         if not two_views:
             seqF = entries[i]

python/obitools3/commands/import.pyx

@@ -109,6 +109,8 @@ def run(config):
     cdef   bint        NUC_SEQS_view
     cdef   bint        silva
     cdef   bint        rdp
+    cdef   bint        unite
+    cdef   bint        sintax
     cdef   int         nb_elts
     cdef   object      d
     cdef   View        view
@@ -234,11 +236,15 @@ def run(config):
         def_col = view[DEFINITION_COLUMN]
         seq_col = view[NUC_SEQUENCE_COLUMN]
 
-    # Prepare taxon scientific name and taxid refs if RDP or SILVA file
+    # Prepare taxon scientific name and taxid refs if RDP/SILVA/UNITE/SINTAX formats
     silva = False
     rdp = False
     unite = False
-    if 'inputformat' in config['obi'] and (config['obi']['inputformat'] == b"silva" or config['obi']['inputformat'] == b"rdp" or config['obi']['inputformat'] == b"unite"):
+    sintax=False
+    if 'inputformat' in config['obi'] and (config['obi']['inputformat'] == b"silva" or \
+                                           config['obi']['inputformat'] == b"rdp" or \
+                                           config['obi']['inputformat'] == b"unite" or \
+                                           config['obi']['inputformat'] == b"sintax"):
         #if taxo is None:
         #    raise Exception("A taxonomy (as built by 'obi import --taxdump') must be provided for SILVA and RDP files")
         if config['obi']['inputformat'] == b"silva":
@@ -247,6 +253,8 @@ def run(config):
             rdp = True
         elif config['obi']['inputformat'] == b"unite":
             unite = True
+        elif config['obi']['inputformat'] == b"sintax":
+            sintax = True
         sci_name_col = Column.new_column(view, SCIENTIFIC_NAME_COLUMN, OBI_STR)
         if taxo is not None:
             taxid_col = Column.new_column(view, TAXID_COLUMN, OBI_INT)
@@ -356,13 +364,26 @@ def run(config):
                     qual_col[i] = entry.quality
                     
                 # Parse taxon scientific name if RDP or Silva or Unite file
-                if (rdp or silva or unite):
+                if (rdp or silva or unite or sintax):
                     if rdp or silva:
                         sci_names = entry.definition.split(b";")
                         sci_name_col[i] = sci_names[-1]
                     elif unite:
                         sci_names = entry.id.split(b'|')[-1].split(b';')
                         sci_name_col[i] = re.sub(b'[a-zA-Z]__', b'', sci_names[-1])
+                    elif sintax:
+                        reconstructed_line = entry.id+b' '+entry.definition[:-1]
+                        splitted_reconstructed_line = reconstructed_line.split(b';')
+                        taxa = splitted_reconstructed_line[1].split(b'=')[1]
+                        taxa = splitted_reconstructed_line[1].split(b',')
+                        sci_names = []
+                        for t in taxa:
+                            tf = t.split(b':')[1]
+                            sci_names.append(tf)
+                        sci_name_col[i] = sci_names[-1]
+                        id_col[i] = reconstructed_line.split(b';')[0]    
+                        def_col[i] = reconstructed_line
+                        
                     # Fond taxid if taxonomy provided
                     if taxo is not None :
                         for sci_name in reversed(sci_names):

python/obitools3/commands/ngsfilter.pyx

@@ -264,6 +264,10 @@ cdef tuple annotate(sequences, infos, no_tags, verbose=False):
     
     not_aligned = len(sequences) > 1
     sequences[0] = sequences[0].clone()
+    
+    if not_aligned:
+        sequences[0][b"R1_parent"] = sequences[0].id
+        sequences[0][b"R2_parent"] = sequences[1].id
 
     if not_aligned:
         sequences[1] = sequences[1].clone()

python/obitools3/parsers/ngsfilter.pyx

@@ -48,24 +48,25 @@ def ngsfilterIterator(lineiterator,
         all_lines.insert(0, firstline)
         
     # Insert header for column names
-    column_names = [b"experiment", b"sample", b"forward_tag", b"reverse_tag", b"forward_primer", b"reverse_primer",b"additional_info"]
+    column_names = [b"experiment", b"sample", b"forward_tag", b"reverse_tag", b"forward_primer", b"reverse_primer"] #,b"additional_info"]
     header = out_sep.join(column_names)
     
     new_lines.append(header)
         
     for line in all_lines:
         split_line = line.split(maxsplit=5)
-        tags = split_line.pop(2)
-        tags = tags.split(b":")
-        for t_idx in range(len(tags)):
-            if tags[t_idx]==b"-" or tags[t_idx]==b"None" or tags[t_idx]==b"":
-                tags[t_idx] = nastring
-        if len(tags) == 1:          # Forward and reverse tags are the same
-            tags.append(tags[0])
-        split_line.insert(2, tags[0])
-        split_line.insert(3, tags[1])
-        new_lines.append(out_sep.join(split_line[0:7]))
-    
+        if split_line:
+            tags = split_line.pop(2)
+            tags = tags.split(b":")
+            for t_idx in range(len(tags)):
+                if tags[t_idx]==b"-" or tags[t_idx]==b"None" or tags[t_idx]==b"":
+                    tags[t_idx] = nastring
+            if len(tags) == 1:          # Forward and reverse tags are the same
+                tags.append(tags[0])
+            split_line.insert(2, tags[0])
+            split_line.insert(3, tags[1])
+            new_lines.append(out_sep.join(split_line[0:6]))
+        
     return tabIterator(iter(new_lines),
                        header = True,
                        sep = out_sep,

python/obitools3/uri/decode.pyx

@@ -506,7 +506,7 @@ def open_uri(uri,
         if format is not None:
             if seqtype==b"nuc":
                 objclass = Nuc_Seq    # Nuc_Seq_Stored? TODO
-                if format==b"fasta" or format==b"silva" or format==b"rdp" or format == b"unite":
+                if format==b"fasta" or format==b"silva" or format==b"rdp" or format == b"unite" or format == b"sintax":
                     if input:
                         iseq = fastaNucIterator(file, 
                                                 skip=skip, 

python/obitools3/version.py

@@ -1,5 +1,5 @@
 major = 3
 minor = 0
-serial= '1b17'
+serial= '1b20'
 
 version ="%d.%d.%s" % (major,minor,serial)

src/.gitignore

@@ -3,3 +3,4 @@
 /cmake_install.cmake
 /libcobitools3.dylib
 /Makefile
+/build/