def new_nexus_without_sites(nexus_obj, sites_to_remove):
"""
Returns a new NexusReader instance with the sites in
`sites_to_remove` removed.
:param nexus_obj: A `NexusReader` instance
:type nexus_obj: NexusReader
:param sites_to_remove: A list of site numbers
:type sites_to_remove: List
:return: A NexusWriter instance
:raises AssertionError: if nexus_obj is not a nexus
:raises NexusFormatException: if nexus_obj does not have a `data` block
"""
check_for_valid_NexusReader(nexus_obj, required_blocks=['data'])
# make new nexus
nexout = NexusWriter()
nexout.add_comment(
"Removed %d sites: %s" %
(len(sites_to_remove), ",".join(["%s" % s for s in sites_to_remove])))
new_sitepos = 0
for sitepos in range(nexus_obj.data.nchar):
if sitepos in sites_to_remove:
continue # skip!
for taxon, data in nexus_obj.data:
nexout.add(taxon, new_sitepos, data[sitepos])
new_sitepos += 1
return nexout
def shufflenexus(nexus_obj, resample=False):
"""
Shuffles the characters between each taxon to create a new nexus
:param nexus_obj: A `NexusReader` instance
:type nexus_obj: NexusReader
:param resample: The number of characters to resample. If set to False, then
the number of characters will equal the number of characters in the
original data file.
:type resample: Integer
:return: A shuffled NexusReader instance
:raises AssertionError: if nexus_obj is not a nexus
:raises ValueError: if resample is not False or a positive Integer
:raises NexusFormatException: if nexus_obj does not have a `data` block
"""
check_for_valid_NexusReader(nexus_obj, required_blocks=['data'])
if resample is False:
resample = nexus_obj.data.nchar
try:
resample = int(resample)
except ValueError:
raise ValueError('resample must be a positive integer or False!')
if resample < 1:
raise ValueError('resample must be a positive integer or False!')
newnexus = NexusWriter()
newnexus.add_comment("Randomised Nexus generated from %s" %
nexus_obj.filename)
for i in range(resample):
# pick existing character
character = randrange(0, nexus_obj.data.nchar)
chars = nexus_obj.data.characters[character]
site_values = [chars[taxon] for taxon in nexus_obj.data.taxa]
shuffle(site_values)
for taxon in nexus_obj.data.taxa:
newnexus.add(taxon, i, site_values.pop(0))
return newnexus
def shufflenexus(nexus_obj, resample=False):
"""
Shuffles the characters between each taxon to create a new nexus
:param nexus_obj: A `NexusReader` instance
:type nexus_obj: NexusReader
:param resample: The number of characters to resample. If set to False, then
the number of characters will equal the number of characters in the
original data file.
:type resample: Integer
:return: A shuffled NexusReader instance
:raises AssertionError: if nexus_obj is not a nexus
:raises ValueError: if resample is not False or a positive Integer
:raises NexusFormatException: if nexus_obj does not have a `data` block
"""
check_for_valid_NexusReader(nexus_obj, required_blocks=['data'])
if resample is False:
resample = nexus_obj.data.nchar
try:
resample = int(resample)
except ValueError:
raise ValueError('resample must be a positive integer or False!')
if resample < 1:
raise ValueError('resample must be a positive integer or False!')
newnexus = NexusWriter()
newnexus.add_comment("Randomised Nexus generated from %s" % nexus_obj.filename)
for i in range(resample):
# pick existing character
character = randrange(0, nexus_obj.data.nchar)
chars = nexus_obj.data.characters[character]
site_values = [chars[taxon] for taxon in nexus_obj.data.taxa]
shuffle(site_values)
for taxon in nexus_obj.data.taxa:
newnexus.add(taxon, i, site_values.pop(0))
return newnexus
def combine_nexuses(nexuslist):
"""
Combines a list of NexusReader instances into a single nexus
:param nexuslist: A list of NexusReader instances
:type nexuslist: List
:return: A NexusWriter instance
:raises TypeError: if nexuslist is not a list of NexusReader instances
:raises IOError: if unable to read an file in nexuslist
:raises NexusFormatException: if a nexus file does not have a `data` block
"""
out = NexusWriter()
charpos = 0
for nex_id, nex in enumerate(nexuslist, 1):
check_for_valid_NexusReader(nex, required_blocks=['data'])
if hasattr(nex, 'short_filename'):
nexus_label = os.path.splitext(nex.short_filename)[0]
elif hasattr(nex, 'label'):
nexus_label = nex.label
else:
nexus_label = str(nex_id)
out.add_comment("%d - %d: %s" %
(charpos, charpos + nex.data.nchar - 1, nexus_label))
for site_idx, site in enumerate(sorted(nex.data.characters), 0):
data = nex.data.characters.get(site)
charpos += 1
# work out character label
charlabel = nex.data.charlabels.get(site_idx, site_idx + 1)
label = '%s.%s' % (nexus_label, charlabel)
for taxon, value in data.items():
out.add(taxon, label, value)
return out
def read_file(self):
with open( self.input_file, 'r') as f:
self.first_line = f.readline()
# TODO: Needs work xread / tnt file format is loose
# xread has some other potential clues
xread_filename = f.readline().strip().replace("'", "")
xread_matrix_dimensions = f.readline()
lines = f.readlines()
f.close
if "#NEXUS" == self.first_line.strip():
print "text file is nexus"
# move to nexus folder
filename, file_extension = os.path.splitext(self.input_file)
os.rename(self.input_file, filename + ".nex")
# send to correct matrix handler
# TODO: A little dirty, probably should
# change mediator's handler param so this flows cleaner
matrixHandler = NexusHandler( filename + ".nex" )
return matrixHandler.read_file()
elif "xread" == self.first_line.strip():
print "xread format found"
dimensions = str(xread_matrix_dimensions).split(' ')
self.ncols = int(dimensions[0])
self.nrows = int(dimensions[1])
custom_block = "\n\nBEGIN VERIFIED_TAXA;\n"
custom_block += "Dimensions ntax=" + str(self.nrows) + " nchar=4;\n"
matrix = ""
matrix_arr = []
line_buffer = ""
row_taxa = []
for l in lines:
if ";proc/;" != l.strip():
if line_buffer:
line_row = line_buffer + " " + l.strip()
else:
line_row = l.strip()
if len(line_row) >= self.ncols:
# reconstitute broken rows, then remove space/tabbing
line_parts = line_row.split(' ')
line_parts = list(filter(None, line_parts))
taxon_name = line_parts[0]
taxon_chars = line_parts[1]
#taxon_chars = line_parts[1].replace("[", "(")
#taxon_chars = taxon_chars.replace("]", ")")
# verify taxa
verified_taxa = verifyTaxa(taxon_name)
verified_name = None
if verified_taxa:
for taxa in verified_taxa:
# We split here to exclude the odd citation on the taxon name ( maybe regex what looks like name & name, year would be better )
verified_name = taxa['name_string'].lower().split(' ')
row_taxa.append( verified_name[0] )
custom_block += taxon_name + " " + taxa['name_string'] + " " + taxa['match_value'] + " " + taxa['datasource'] + "\n"
matrix += " " + verified_name[0] + " " + taxon_chars.strip() + "\n"
matrix_arr.append(taxon_chars.strip())
else:
row_taxa.append( taxon_name )
custom_block += taxon_name + "\n"
matrix += " " + taxon_name + " " + taxon_chars.strip() + "\n"
matrix_arr.append(taxon_chars.strip())
line_buffer = ""
else:
line_buffer += l.strip()
custom_block += ";\n"
custom_block += "END;\n"
self.custom_block = custom_block
print "matrix array"
marr = []
for row in matrix_arr:
items = list(row)
marr.append(items)
m = numpy.matrix(marr)
nw = NexusWriter()
nw.add_comment("Morphobank generated Nexus from xread .txt file ")
for rx in range(self.nrows):
taxon_name = row_taxa[rx]
cell_value = m.item(rx)
for cindex, cv in enumerate(cell_value):
char_no = cindex + 1
nw.add(taxon_name, char_no, cv)
# keep the upload path and filename, but change extension
file_parts = self.input_file.split('.')
output_file = file_parts[0] + '.nex'
nw.write_to_file(filename=output_file, interleave=False, charblock=True)
# move to nexus folder
#os.rename(xread_filename + ".nex", "./nexus/" + xread_filename + ".nex")
# wait for file to move before open and append
#while not os.path.exists('./nexus/' + xread_filename + '.nex'):
# time.sleep(1)
#if os.path.isfile('./nexus/' + xread_filename + '.nex'):
# Custom Block Section
nexus_file = codecs.open(output_file, 'a', 'utf-8')
nexus_file.write(custom_block)
nexus_file.close()
return output_file
else:
print "do not know how to process this .txt file"
def read_file(self):
book = xlrd.open_workbook(self.input_file)
sh = book.sheet_by_index(0)
nw = NexusWriter()
self.nrows = sh.nrows
self.ncols = sh.ncols
nw.add_comment("Morphobank generated Nexus File")
custom_block = "\n\nBEGIN VERIFIED_TAXA;\n"
custom_block += "Dimensions ntax=" + str(self.nrows) + " nchar=4;\n"
# Species List Taxa
for rx in range(self.nrows):
if rx:
taxon_name = str(sh.cell_value(rowx=rx, colx=0)).strip()
verified_taxa = verifyTaxa(taxon_name)
verified_name = None
if verified_taxa:
for taxa in verified_taxa:
verified_name = taxa['name_string'].lower()
custom_block += taxon_name + " " + taxa['name_string'] + " " + taxa['match_value'] + " " + taxa['datasource'] + "\n"
else:
custom_block += taxon_name + "\n"
for cx in range(self.ncols):
if cx:
if is_number( sh.cell_value(rowx=rx, colx=cx)):
cell_value = int(sh.cell_value(rowx=rx, colx=cx))
else:
cell_value = sh.cell_value(rowx=rx, colx=cx)
cell_value = cell_value.replace("{", "(")
cell_value = cell_value.replace("}", ")")
if verified_name:
nw.add(verified_name, cx, cell_value)
else:
nw.add(taxon_name, cx, cell_value)
custom_block += ";\n"
custom_block += "END;\n"
self.custom_block = custom_block
# keep the upload path and filename, but change extension
file_parts = self.input_file.split('.')
output_file = file_parts[0] + '.nex'
nw.write_to_file(filename=file_parts[0] + '.nex', interleave=False, charblock=True)
## quick append Custom Block
nexus_file = open(output_file, 'a')
nexus_file.write(custom_block)
nexus_file.close()
return output_file
