def Linnaeus2YASDBF(params):
db_name = params['DB Name']
linnaeus = '/home/d3y034/chembio_svn/Proteomics/Linnaeus/Linnaeus.py'
if os.path.exists(linnaeus):
yasdbf_main.runCmd( "%s %s %s" % (linnaeus, 'compile2YASDBF', db_name) )
else:
sys.exit('Linnaeus is not in the specifed location.')
def insertSpectraSTData(con, pepXML):
# NOTE: consider avoiding hardcoding this
spectraST = "/usr/bin/spectrast"
if os.path.exists(spectraST):
# keep relative path in base
path, base = os.path.split(pepXML)
base = base.split('.')[0]
yasdbf_main.runCmd( "%s -c -cP0 %s" % (spectraST, pepXML) )
sptxt = os.path.join(path, base + '.sptxt')
specST_dict = parseSPTXT(sptxt)
# NOTE: try to alter table
# TODO: remove!
try:
alter_table(con)
except sqlite3.OperationalError, ex:
if "duplicate column name" in str(ex):
pass
else:
raise
data = []
for k, v in specST_dict.iteritems():
data.append( (';'.join(v[0]), ';'.join(v[1]) , ';'.join(v[2]), v[3], v[4], v[5], k) )
con.executemany("UPDATE spectra SET x=?, y=?, labels=?, precur_int=?, total_int=?, matched_int=? WHERE name=?", data)
con.commit()
def runMSGF(msgfPre, dirname, enzyme):
if os.path.exists(msgfPre):
msgfPath = "/home/yasdbf/apps/msgf.jar"
msgf = os.path.join(dirname, msgfPre[:-3]) #make this a path
if len(dirname) is 0:
dirname = "."
params = (msgfPath, msgfPre, dirname, msgf, enzyme)
cmdStr = "java -Xmx2000M -jar %s -i %s -d %s -o %s -fixMod 0 -e %s" % params
# "-fixMod 0" sets default mods to none
# execute command
yasdbf_main.runCmd(cmdStr)
return msgf
#else
return None
def Tandem2XML(tandemXML):
Tandem2XML = "/home/d3y034/chembio_svn/Proteomics/Kraftwerk/apps/Tandem2XML"
if os.path.exists(Tandem2XML):
d = os.path.dirname(tandemXML)
base = os.path.basename(tandemXML).split('.')[0]
# NOTE: need to copy tandem file to new name for it to relate to mzXML
newTandemXML = os.path.join(d, base + '.xml')
yasdbf_main.runCmd( "cp %s %s" % (tandemXML, newTandemXML) )
# NOTE: this means pepXMLs follow naming of pepXMLs
pepXML = os.path.join(d, base + '.pep.xml')
yasdbf_main.runCmd( "%s %s %s" % (Tandem2XML, newTandemXML, pepXML) )
# NOTE: remove copy
yasdbf_main.runCmd( "rm %s" % newTandemXML )
return pepXML
else:
sys.exit("Tandem2XML does not appear to be installed")
if "duplicate column name" in str(ex):
#NOTE: maybe log some kind of error here
pass
else:
raise
enzyme = params['Enzyme']
filename = filename.split('.')[0] + '.mzXML'
filename = os.path.basename(filename) # make sure you have basename (important for structured mode)
file_id, dirname = con.execute("SELECT id, dirname FROM files WHERE basename=?", (filename,) ).fetchone()
con.row_factory = sqlite3.Row
msgfPre = writePrefile(con, file_id, filename)
msgf = runMSGF(msgfPre, dirname, enzyme)
yasdbf_main.runCmd("rm %s" % msgfPre) # clean-up
importMSGF(con, msgf)
def writePrefile(con, file_id, basename):
# use basename to find name for msgfPre file
basename = basename.split('.')[0] + '.mzXML'
filename = basename.split('.')[0] + '.msgfPre'
# write msgfPre file
with open(filename, 'w') as f:
# write headers
f.write("#SpectrumFile\tScanName\tScan#\tAnnotation\tCharge\n")
select = """SELECT name, n_terminal, peptide, peptide_modded, c_terminal, charge
FROM spectra WHERE file_id = ? ORDER BY name"""
for record in con.execute(select, (file_id,)):