def get_cluster_elements(self):
# need to search for elements in the fasta file
# going to assume that headers is always made to the first space
fasta = self.parent_file.split('.')[0]
# fasta duplicate always made by cd hit which has the same
# file name but no extension as the clstr file
try:
lines = []
fasta_elements = []
search_dict = {}
with open(self.parent_file) as parent:
lines = read_as_tuples(parent)
for element_tuple in lines:
search_dict[element_tuple[0].split(' ')[0]] = element_tuple
# use up to first space as key
for clstr_element in self.elements:
if clstr_element.name in search_dict:
fasta_elements.append(search_dict[clstr_element.name])
# adds all elements in a cluster to the fasta elements list
return fasta_elements # list of tuples
except FileNotFoundError as e:
return e
def write_cluster_fastas(self, original_fasta_path, path):
'''
Iterates through all clusters in the cluster set and creates a fasta
file of the elements in those clusters. Path is where new dir containing
a cluster will be written to. The intention being that then can use
consensus_tools to create consensus seqs from each of the cluster
files. Returns the path files are written to.
'''
search_dict = {} # assumes full path
lines = read_as_tuples(original_fasta_path)
# all clusters of a file from the same original fasta
for element_tuple in lines:
search_dict[element_tuple[0].split(' ')[0]] = element_tuple
fasta_paths = [] # store the paths of written files
for cluster in self.clusters_set:
write_list = [] # contains elements in cluster to be written to file
# iterate through all clusters
file_name = os.path.join(path, '{}_cluster_{}'.format(
os.path.basename(cluster.parent_file.split('.')[0]), cluster.num.strip()))
fasta_paths.append(file_name)
count = 0
for clstr_element in cluster.elements:
if clstr_element.name in search_dict:
write_list.append(search_dict[clstr_element.name])
# write all elements found in dictionart to cluster fasta file
write_from_tuple_list(write_list, file_name)
cluster.fasta = file_name # change fasta variable of the cluster
return fasta_paths
def convert_rep_seqs_to_longest_orf(rep_seq_fasta):
'''
Given a fasta file of sequnces (should be the rep sequences) returns a list
of tuples where first item is the oringal header and second is the longest
ORF found in that header's sequence. Can then be passed off to the tblastn
search method.
'''
rep_proteins = []
for rep_header, rep_seq in read_as_tuples(rep_seq_fasta):
rep_proteins.append(
(rep_header, find_longest_orfs(six_frames(rep_seq))))
write_from_tuple_list(rep_proteins, rep_seq_fasta)
return rep_proteins # returns as a list of tuples
def get_rep_sequence(fasta_file_path, clstr_file_path):
'''
Given a path to a fasta file and the clstr file produced by cd hit for the
fasta file returns the header and sequence of the representative sequence
identified by cd-hit as a tuple.
'''
fasta_files = read_as_tuples(fasta_file_path)
target = format_rep_header(get_largest_cluster_rep(clstr_file_path))
# pulls out a name that looks like >name=RLC_Gmr2_Gm1-29 which will be
# the first thing in the header of the seq in the fasta file we want to
# match to
for header, seq in fasta_files:
name = header.split(' ')[0]
# gives just the name of the element
# assuming standard soybase fasta header format
if name == target:
return header, seq
return False # no match found uh oh :(
def get_cluster_elements(self):
# need to search for elements in the fasta file
# going to assume that headers is always made to the first space
fasta = self.parent_file.split('.')[0]
# fasta duplicate always made by cd hit which has the same
# file name but no extension as the clstr file
try:
lines = []
fasta_elements = []
search_dict = {}
lines = read_as_tuples(self.parent_file.split('.')
[0]) # read as tuples opens the file
for element_tuple in lines:
search_dict[element_tuple[0].split(' ')[0]] = element_tuple
except FileNotFoundError as e:
return e