def armi():
"""Performs the necessary analyses on strains using armi targets"""
global targetpath, seqdict
# Set the analysis type
analysistype = "armi"
# Set the path of the target files
currenttargetpath = "%s%s" % (targetpath, analysistype)
# Create the bait files as necessary
baittargets(currenttargetpath, analysistype)
# The cat file is all the target files concatenated together
catfile = "%s/%sConcatenated.fasta" % (currenttargetpath, analysistype)
# If the cat file doesn't exist, create it from the bait file generated in baittargets()
if not os.path.isfile(catfile):
shutil.copyfile("%s/bait/%sBait.fa" % (currenttargetpath, analysistype), catfile)
# In order to save time, a precomputed hash file is used
hashfile = glob("%s/bait/*.gz" % currenttargetpath)
# If this precomputed hash exists, use it
if hashfile:
baitfile = hashfile[0]
# Otherwise use the .fasta bait file
else:
baitfile = glob("%s/bait/*.fa*" % currenttargetpath)[0]
# Get the full target database into a variable
armidatabase = glob("%s/*.fa*" % currenttargetpath)
# Filter faidx processed targets from the list
armidatabase = [target for target in armidatabase if ".fai" not in target and "Concatenated" not in target]
# Add necessary variables to seqdict
for folderName in seqdict:
seqdict[folderName]["bait"]["fastqFiles"][analysistype] = baitfile
seqdict[folderName]["targets"][analysistype] = armidatabase
seqdict[folderName]["concatenatedTargets"][analysistype] = catfile
print "Filtering .fastq files with %s targets" % analysistype
# Run the baiting process
baitrprocesses(analysistype)
print "\nIndexing %s targets" % analysistype
# Index the combined target file
SMALTcombined.SMALTindexTargets(catfile, currenttargetpath)
print '\nPerforming %s reference mapping' % analysistype
# Use SMALT to perform reference mapping of the combined target file
SMALTcombined.SMALTmappingProcesses(seqdict, analysistype, "SMALT")
print "\nSorting mapped %s files" % analysistype
# Use samtools to sort the bam file
bamProcessorCombined.sortingprocesses(seqdict, analysistype)
print '\nIndexing sorted %s files' % analysistype
# Use samtools to index the sorted bam file
bamProcessorCombined.bamindexingprocesses(seqdict, analysistype)
print '\nParsing %s results' % analysistype
# Use pysamstats to parse the bam files. Mike's armi module is called from within bamPysamStatsCombined
bamPysamStatsCombined.bamParseProcesses(seqdict, analysistype, reportfolder)
def pathotyper(organismdict, organismlist, analysistype):
"""
Performs the necessary analyses on strains using genus-specific pathotype/serotype targets
:param organismdict: dictionary of the 16S results
:param organismlist: list of all the genera in the current analysis
:param analysistype: string of the current analysis type
"""
global targetpath, seqdict
# Targets are stored in targetpath/Organism/<genus>/<analysistype>/
currenttargetpath = "%sOrganism" % targetpath
# Print this prior to the loop
print '\nIndexing %s target files' % analysistype
# As strains will be processed depending differently based on their genus, they must be processed separately
for strain in organismdict:
# Try/except account for genera without pathotyping schemes
try:
# Set the target path
pathopath = glob("%s/%s/%s" % (currenttargetpath, organismdict[strain].keys()[0], analysistype))[0]
# Create the bait targets if necessary
baittargets(pathopath, analysistype)
# The cat file will be used in the "combined" reference mapping
catfile = "%s/%sConcatenated.fasta" % (pathopath, analysistype)
# If the cat file does not exist, copy the bait file from the bait folder to the target folder
if not os.path.isfile(catfile):
shutil.copyfile("%s/bait/%sBait.fa" % (pathopath, analysistype), catfile)
# Find the files to be used in baiting - if a precomputed hash file is present, use it
hashfile = glob("%s/bait/*.gz" % pathopath)
if hashfile:
baitfile = hashfile[0]
else:
baitfile = glob("%s/bait/*.fa*" % pathopath)[0]
# Set the baittype variable as the genus, and the analysis type
baittype = "%s_%s" % (organismdict[strain].keys()[0], analysistype)
# Get all the targets into a list
# Even though the cat file will be used for the reference mapping rather than the individual target files,
# the target names (taken from the name of the target files) is still used in the parsing of results
targets = glob("%s/*.fa*" % pathopath)
# Remove faidx processed files from the list
targets = [target for target in targets if ".fai" not in target and "Concatenated" not in target]
# Add the bait file, the cat file, and the list of targets to seqdict
seqdict[strain]["bait"]["fastqFiles"][baittype] = baitfile
seqdict[strain]["targets"][analysistype] = targets
seqdict[strain]["concatenatedTargets"][analysistype] = catfile
# Index the SMALT targets
SMALTcombined.SMALTindexTargets(catfile, pathopath)
except IndexError:
pass
# Bait!
print "Filtering .fastq files with %s targets" % analysistype
baitrprocesses(analysistype)
print '\nPerforming reference mapping'
# Use SMALT to perform reference mapping
SMALTcombined.SMALTmappingProcesses(seqdict, analysistype, "SMALT")
print '\nSorting mapped %s files' % analysistype
# Use samtools to sort bam files
bamProcessorCombined.sortingprocesses(seqdict, analysistype)
print '\nIndexing sorted %s files' % analysistype
# Use samtools to index sorted bam files
bamProcessorCombined.bamindexingprocesses(seqdict, analysistype)
print '\nParsing %s results' % analysistype
# Use pysamstats to parse results
pathomatches = bamPysamStatsCombined.bamParseProcesses(seqdict, analysistype, reportfolder)
# Create a report
pathoreportr(pathomatches, analysistype, organismdict, organismlist)