ret = cfg.evaluateCommandLineArgs(sys.argv)
cfg.scalePpmMda()
dataDir = cfg.parameters['runtime']['datadir']
fileFilter = cfg.parameters['runtime']['filefilter']
hcdOnly = cfg.parameters['general']['hcdonly']
logParam = cfg.parameters['logging']
logPath = Path(dataDir.joinpath(logParam['logdir']))
if not logPath.exists():
logPath.mkdir(parents=True)
logFile = logPath.joinpath(logParam['logfile'])
logger = Logger(logFile, logParam['loglevel'], logParam['screenlevel'],
False)
logger.setLog('mgfCreation')
logger.log.info('Analysing: %s' % str(dataDir.joinpath(fileFilter)))
if hcdOnly:
logger.log.info('Exporting HCD data only')
# for f in dataDir.files(fileFilter):
for f in dataDir.glob(fileFilter):
if not f.is_file():
# skip any directories
continue
# if f.name[:4] in ['6528', '1814', '2032']: continue
mgf = mgftools(f)
logger.log.info('Filename: %s' % f.name)
if hcdOnly:
if __name__ == "__main__":
logger = 0
cfg = ConfigManager('./corrects2iquant.cfg')
try:
configErr = cfg.evaluateCommandLineArgs(sys.argv)
dataDir = cfg.parameters['runtime']['datadir']
# start the logging process
logParam = cfg.parameters['logging']
logPath = Path(dataDir.joinpath(logParam['logdir']))
if not logPath.exists():
logPath.mkdir(parents=True)
logFile = logPath.joinpath(logParam['logfile'])
logger = Logger(logFile, logParam['loglevel'], logParam['screenlevel'], False)
logger.setLog('corrects2iquant')
cfg.log = logger.log
hdf5files = [Path(cfg.parameters['runtime']['datadir'], cfg.parameters['runtime']['filename'])]
allpeptideratios = {}
allsumionratiodata = {}
corrects2iquantoblist = []
counter = 0
for hdf5Path in hdf5files:
logger.log.info('hdf5Path %s' % hdf5Path)
logger.log.info('starting')
hdf5corrects2iquant = HDF5Results(hdf5Path)
hdf5corrects2iquant.appendOpen()
hdf5corrects2iquant.samplename = hdf5Path.name
hdf5corrects2iquant.cfg = cfg
hdf5corrects2iquant.addConfigParameters(cfg.parameters, 'postMascot', '4 corrects2iquant')
dataDir = cfg.parameters['runtime']['datadir']
pyMSsafeAppDir = str(Path('./pyMSsafe').resolve())
logParam = cfg.parameters['logging']
logPath = Path(dataDir.joinpath(logParam['logdir']))
if logPath.exists():
# clear existing logs for this part of the pipeline
for log in logPath.glob('*.log'):
if log.name in ['preMascot.log', 'pyMSsafe.log', 'mgfCreation.log']:
log.unlink()
else:
logPath.mkdir(parents=True)
logFile = logPath.joinpath(logParam['logfile'])
logger = Logger(logFile, logParam['loglevel'], logParam['screenlevel']) # , pID)
logger.setLog('preMascot')
logger.log.log(logger.PROCESS, 'started preMascot application')
logger.log.info('Removing previous errors from %s' % str(dataDir))
# clear previous error files
for errorFile in dataDir.glob('*.error'):
errorFile.unlink()
logger.log.info('Analysing: %s' % str(dataDir))
rawFileList = []
for idx, raw in enumerate(dataDir.glob('*.raw')):
rawFileList.append((raw, idx + 1))
logger.log.info('Found %i files to process' % len(rawFileList))
if not rawFileList:
logger.log.critical('Found %i files to process' % len(rawFileList))
pyMSsafeDir = Path('./pyMSsafe').absolute()
class pymssafe:
def __init__(self, config):
"""
@brief initiaise pyMSsafe controller
@param mode <string>: operational mode of the application
"""
self.cfg = config
dataDir = cfg.parameters['runtime']['datadir']
logParam = cfg.parameters['logging']
logPath = Path(dataDir.joinpath(logParam['logdir']))
if not logPath.exists():
logPath.mkdir(parents=True)
logFile = logPath.joinpath(logParam['logfile'])
self.logs = Logger(logFile, logParam['loglevel'],
logParam['screenlevel'], False)
self.logs.setLog('pyMSsafe')
def run(self, jobObj):
"""
@brief run the analysis
@param job <object>: containing all the job data
"""
# needs to catch exceptions here so that LUX doesn't recieve an exception
hdf = ''
xraw = ''
tempFile = ''
config = self.cfg
logs = self.logs
try:
hdf = ''
maxspec = jobObj.args['maxspec']
logs.log.info("Starting PyMSSafe runner")
# assign objects to the file interfaces
rawpath = jobObj.srcpth.absolute()
namebase = rawpath.stem
hdfpath = jobObj.dstpth.absolute().joinpath(namebase + '.hdf5')
runname = namebase
tempFile = jobObj.dstpth.absolute().joinpath(
str(config.parameters['runtime']['pid']))
if tempFile.exists():
tempFile.unlink()
logs.log.info('started job: %s' % rawpath.name)
watch = Stopwatch()
if not rawpath.exists():
logs.log.info('could not find file: %s' % str(rawpath))
logs.log.info('Stopped')
return {
'code': 1,
'error': 'could not find file: %s' % str(rawpath)
}
# get quant information if file is quan type
quantMeth = self.extractQuant(jobObj.srcpth.name.upper())
if quantMeth == -1:
raise ExHa.QuantificationMethodError(
'Unable to find "%s" quantification method' %
self.cfg.parameters['runtime']['quant'].upper())
elif quantMeth == -2:
raise ExHa.QuantificationMethodError(
'Unable to find valid quantification method in file name (%s)'
% jobObj.srcpth.name.upper())
xraw = XRawFile(str(rawpath))
if config.parameters['general']['skipscanevents']:
xraw.skipScanEvents = config.parameters['general'][
'skipscanevents']
# opens hdf5 file for writing
hdf = hdf5Base(hdfpath, True, True)
hdf.appendOpen()
self.createHDFtables(hdf)
# save the config entries
hdf.appendRows('/rawdata/config', config.convertConfig())
# create datamanager object
config.rawfile = jobObj.srcpth
dataman = Datamanager(xraw, config, logs, hdf, quantMeth,
str(tempFile))
dataman.maxspec = maxspec
dataman.addLUXdata(jobObj, config.parameters['runtime']['pid'])
# run the analysis in the datamanager
ok = dataman.processSpectra(quantMeth, watch, runname)
if ok['code'] != 0:
raise ExHa.SpectraProcessingError(ok['error'])
# run the XIC generation
logs.log.info('Processing XIC data for %d MS/MS events' %
len(dataman.specs))
ok = dataman.processXICs()
if ok['code'] != 0:
raise ExHa.XICprocessingError(ok['error'])
watch.rec('Processing XIC')
logs.log.info('Writing HDF5 indexes')
hdf.indexTable('/rawdata/msmsheader', ['spec_id'])
hdf.close()
xraw = ''
hdf = ''
watch.stop()
logs.log.info('job took: %s' % watch.oneLineFormat())
tempFile.unlink()
except ExHa.MSdataConsistancyError, msg:
self.shutdown(hdf, xraw, tempFile)
logs.log.warning('error: pyMSsafe Data error: %s' % msg)
return {'code': 2, 'error': 'pyMSsafe Data error: %s' % msg}
except ExHa.XICprocessingError, msg:
self.shutdown(hdf, xraw, tempFile)
logs.log.warning('error: pyMSsafe XIC error: %s' % msg)
return {'code': 3, 'error': 'pyMSsafe XIC error: %s' % msg}
if __name__ == '__main__':
configPath = './outputResults.cfg'
cfg = ConfigManager(configPath)
ret = cfg.evaluateCommandLineArgs(sys.argv)
pID = os.getpid()
dataDir = cfg.parameters['runtime']['datadir']
logParam = cfg.parameters['logging']
logPath = Path(dataDir.joinpath(logParam['logdir']))
if not logPath.exists():
logPath.mkdir(parents=True)
logFile = logPath.joinpath(logParam['logfile'])
logger = Logger(logFile, logParam['loglevel'], logParam['screenlevel'],
False)
logger.setLog('outputResults')
logger.log.info('started outputResults application')
filefiltervalue = cfg.parameters['runtime']['filefilter']
# this weird construct makes sure that we only try and get outputs from .hdf5 files
# if hdf5 alrady in filefilter name then it's removed and readded, otherwise it's added
filefiltervalue = filefiltervalue.replace('.hdf5', '') + '.hdf5'
for hdf in dataDir.glob(filefiltervalue):
outputResults(hdf)
logger.log.info('finished outputResults')
dataDir = cfg.parameters['runtime']['datadir']
logParam = cfg.parameters['logging']
logPath = Path(dataDir.joinpath(logParam['logdir']))
if logPath.exists():
# clear existing logs for this part of the pipeline
for log in logPath.glob('*.log'):
if log.name in ['postMascot.log', 'mascotParser.log', 'proteinQuant.log',
'correctS2Iquant.log', 'output.log']:
log.unlink()
else:
logPath.mkdir(parents=True)
logFile = logPath.joinpath(logParam['logfile'])
logger = Logger(logFile, logParam['loglevel'], logParam['screenlevel'], pID)
logger.setLog('postMascot')
logger.log.log(logger.PROCESS, 'started postMascot application')
logger.log.info('Removing previous errors from %s' % str(dataDir))
# clear previous error files
for errorFile in dataDir.glob('*.error'):
errorFile.unlink()
logger.log.info('Analysing: %s' % str(dataDir))
foundFiles, missingFiles = checkFilePresence(dataDir)
startProcessing = False
if not foundFiles:
logger.log.warning('No *.dat files found for processing. Please make sure these are in the same directory '
'as the .hdf5 files from the preMascot workflow.\nAborting.')
if __name__ == "__main__":
logger = 0
cfg = ConfigManager('./proteinquantification.cfg')
try:
configErr = cfg.evaluateCommandLineArgs(sys.argv)
dataDir = cfg.parameters['runtime']['datadir']
# start the logging process
logParam = cfg.parameters['logging']
logPath = Path(dataDir.joinpath(logParam['logdir']))
if not logPath.exists():
logPath.mkdir(parents=True)
logFile = logPath.joinpath(logParam['logfile'])
logger = Logger(logFile, logParam['loglevel'], logParam['screenlevel'],
False)
logger.setLog('proteinQuantification')
cfg.log = logger.log
hdf5files = [
Path(cfg.parameters['runtime']['datadir'],
cfg.parameters['runtime']['filename'])
]
counter = 0
# go through all results.hdf5 files in list: normally we should only have one for merged datasets
for hdf5Path in hdf5files:
logger.log.info('hdf5Path %s' % hdf5Path)
logger.log.info('starting')
hdf5quanprot = HDF5Results(hdf5Path)
hdf5quanprot.appendOpen()
hdf5quanprot.createQuantTables()