def getAlignment( align_node, bedseq, comm_struct):
name = CommStruct.getAttribute( align_node, BedSeqAlignmentStatsCommStruct.ALIGNMENT_NAME_ATT)
seq_align = SequenceAlignment()
seq_align.name = name
seq_align.referenceSpecies = bedseq.species
comm_struct.addSequenceAlignment( bedseq, seq_align)
return seq_align
def generateRandomMSA(self, msa_length, bedseq_number, max_length,
output_commstruct):
# Retrieve method required parameters
RSAT_PATH = self.component.getParameter(Constants.RSAT_DIR_PARAM)
dir_path = os.path.join(self.component.outputDir,
self.component.getComponentPrefix())
file_path = os.path.join(dir_path, "random_sequences.txt")
try:
# Execute the RSAT random-seq command
cmd = os.path.join(RSAT_PATH, "perl-scripts/random-seq")
cmd += " -l " + str(int(max_length * 1.5))
cmd += " -n " + str(bedseq_number)
cmd += " -a a:t 0.3 c:g 0.2"
cmd += " -type DNA"
cmd += " -format multi"
cmd += " -o " + file_path
Log.info(
"GenerateMSAProcessor.generateMSA : starting random sequence generation. Command used is : "
+ cmd)
# Execute the command
cmd_result = commands.getstatusoutput(cmd)
if cmd_result[0] != 0:
Log.log(
"GenerateMSAProcessor.generateMSA : status returned is :" +
str(cmd_result[0]) + " for command '" + cmd + "'")
Log.log(
"GenerateMSAProcessor.generateMSA : command output is = \n"
+ str(cmd_result[1]))
raise ExecutionException(
"GenerateMSAProcessor.generateMSA : Cannot execute random-seq commands. See logs for more details"
)
# Read the output file to get the random sequences
sequence_list = []
sequence_file = open(file_path, "r")
for line in sequence_file:
sequence_list.append(line.split()[0])
# Generate the species list
species_list = []
species_list.append(output_commstruct.baseSpecies)
for index in range(msa_length - 1):
species_list.append("Species" + str(index + 1))
# Create and fill the MSA for each BED sequence
count_seq = 0
for chrom in output_commstruct.bedSequencesDict.keys():
for bedseq in output_commstruct.bedSequencesDict[chrom]:
msa = SequenceAlignment()
msa.name = bedseq.name + "_1"
msa.referenceSpecies = output_commstruct.baseSpecies
seq_length = bedseq.indexEnd - bedseq.indexStart
sequence = list(sequence_list[count_seq][:seq_length])
for index in range(msa_length):
msa.addSequence(species_list[index], sequence)
#msa.addSequence( species_list[index], list(['.'] * len( sequence)))
msa.finalizeSequences()
output_commstruct.addSequenceAlignment(bedseq, msa)
count_seq += 1
except IOError, io_exce:
raise ExecutionException(
"GenerateMSAProcessor.generateMSA : Unable to save/read random sequences file. From:\n\t---> "
+ str(io_exce))