def convert_quantification(input_integrals_filename, workflow_parameters,
output_filename):
params_obj = proteosafe.parse_xml_file(workflow_parameters)
mangled_mapping = proteosafe.get_mangled_file_mapping(params_obj)
all_input_filename = [
os.path.basename(mangled_mapping[key]) for key in mangled_mapping
]
filename_mapping = {}
for filename in all_input_filename:
removed_extension = os.path.splitext(filename)[0]
filename_mapping[removed_extension] = filename
integrals_df = pd.read_csv(input_integrals_filename, skiprows=[1, 2, 3])
feature_to_rt_mapping = load_feature_to_rt_mapping(
input_integrals_filename)
all_molecules = list(integrals_df.keys())
all_molecules.remove("No:")
output_list = []
for molecule in all_molecules:
output_dict = {}
output_dict["row ID"] = molecule
output_dict["row m/z"] = "0"
output_dict["row retention time"] = feature_to_rt_mapping[molecule]
for record in integrals_df.to_dict(orient="records"):
sample_name = record["No:"]
abundance = record[molecule]
if sample_name in filename_mapping:
sample_name = filename_mapping[sample_name]
output_dict[sample_name + " Peak area"] = abundance
output_list.append(output_dict)
pd.DataFrame(output_list).to_csv(output_filename, sep=",", index=False)
def main():
parser = argparse.ArgumentParser(
description='Processing and feature detecting all gc files')
parser.add_argument('proteosafe_parameters', help='proteosafe_parameters')
parser.add_argument('spectrum_folder', help='spectrum_folder')
parser.add_argument('scratch_folder', help='scratch_folder')
parser.add_argument('clustered_mgf', help='scratch_folder')
parser.add_argument('clusterinfo', help='scratch_folder')
parser.add_argument('clustersummary', help='scratch_folder')
parser.add_argument('summary_output', help='summary_output')
parser.add_argument('python_runtime', help='python_runtime')
parser.add_argument('-import_script',
help='import_script',
default="./proc/io/importmsdata.py")
parser.add_argument('-align_script',
help='align_script',
default="./proc/preproc/intrapalign.py")
parser.add_argument('-noise_script',
help='noise_script',
default="./proc/preproc/noisefilter.py")
parser.add_argument('-interalign_script',
help='interalign_script',
default="./proc/preproc/interpalign.py")
parser.add_argument('-peakdetect_script',
help='peakdetect_script',
default="./proc/preproc/peakdetect.py")
parser.add_argument('-export_script',
help='export_script',
default="./proc/io/export.py")
parser.add_argument('-report_script',
help='report_script',
default="./proc/io/report.py")
parser.add_argument('-vistic_script',
help='vistic_script',
default="./proc/vis/vistic.py")
args = parser.parse_args()
workflow_params = proteosafe.parse_xml_file(args.proteosafe_parameters)
mangled_mapping = proteosafe.get_mangled_file_mapping(workflow_params)
file_type_of_import = determine_filetype_of_import(args.spectrum_folder)
hdf5_filename = os.path.join(args.scratch_folder, "data.h5")
cmds = []
#import data
if workflow_params['TIME_UNIT'][0] == "MIN":
cmds.append([
args.python_runtime, args.import_script, "-f", file_type_of_import,
args.spectrum_folder, hdf5_filename, "--timeunits", "'min'"
])
else:
cmds.append([
args.python_runtime, args.import_script, "-f", file_type_of_import,
args.spectrum_folder, hdf5_filename, "--timeunits", "'sec'"
])
#intra align
cmds.append([
args.python_runtime, args.align_script, hdf5_filename, "--h5writepath",
"'sp2D'"
])
#noisefilter
cmds.append([
args.python_runtime, args.noise_script, hdf5_filename, "--h5readpath",
"'sp2D'", "--h5writepath", "'spproc2D'", "--window", "6", "--frame",
"50"
])
#inter align
cmds.append([
args.python_runtime, args.interalign_script, hdf5_filename,
"--h5readpath", "'spproc2D'", "--h5writepath", "'spal2D'"
]) #According to ipython
#peak detect
cmds.append([
args.python_runtime, args.peakdetect_script, hdf5_filename,
"--h5readpath", "'spal2D'", "--individual", "no", "--frag_pattern",
"deconvolution"
])
#output detection
cmds.append([
args.python_runtime, args.export_script, hdf5_filename,
args.scratch_folder
])
#Additional outputs
#python3 ./mshub/proc/io/export.py ./test.h5 --export_integral_table "yes" --export_ms_peak_list "yes" --logfile ./test.log --overwrite_logfile 'no'
#outputting TIC with viz package
#cmds.append([args.python_runtime, args.vistic_script, "--outputfile", os.path.join("..", args.tic_html), "--display", "no", hdf5_filename])
cmds.append([
args.python_runtime, args.report_script, "--output_prefix", "gnps-gc",
hdf5_filename, "summary_temp"
])
cmds.append([
"tar", "-cvf",
os.path.join(args.summary_output, "summary.tar"),
os.path.join("summary_temp", "gnps-gc")
])
for cmd in cmds:
print(" ".join(cmd))
subprocess.call(cmd)
#Parse files
output_peak_txt_filename = os.path.join(args.scratch_folder,
"data_ms_peaks.txt")
output_quant_filename = os.path.join(args.scratch_folder,
"data_integrals.csv")
# mapping the input spec names
mangled_mapping_filename = {}
for key, value in mangled_mapping.items():
mangled_mapping_filename[key.split('.')[0]] = value.split(
'.')[0].split("/")[-1]
f = open(output_quant_filename, 'r').readlines()
quant_in_memory = []
for line in f:
fname = line.split(',')[0]
if fname.startswith("spec") and fname in mangled_mapping_filename:
print("before:")
print(line)
line = line.replace(fname, mangled_mapping_filename[fname])
print("after:")
print(line)
quant_in_memory.append(line)
rewrite_quant = open(output_quant_filename, 'w')
rewrite_quant.write("".join(quant_in_memory))
rewrite_quant.close()
parse_peaks_for_output(output_peak_txt_filename, args.clustered_mgf)
simple_presence_of_merged_spectra_processing(output_quant_filename,
args.clusterinfo,
mangled_mapping)
generate_clustersummary(output_quant_filename, args.clustersummary)
def process_candidate_molecules(candidate_molecules, path_to_spectrum_files, proteosafe_param):
#Grouping by filename
structures_by_filename = defaultdict(list)
for candidate_object in candidate_molecules:
filename = candidate_object["filename"]
structures_by_filename[filename].append(candidate_object)
output_dict = defaultdict(list)
#Demangle
if proteosafe_param:
workflow_params = proteosafe.parse_xml_file(proteosafe_param)
mangled_mapping = proteosafe.get_mangled_file_mapping(workflow_params)
reversed_mapping = {}
for key, value in mangled_mapping.items():
fn = value.split("/")[-1]
reversed_mapping[fn]=key
for filename in structures_by_filename:
#print(filename)
# if param exists => proteosafe workflow => demangle fileName
path_to_spectrum_file = os.path.join(path_to_spectrum_files, filename)
displaying_filename = filename
if proteosafe_param:
#This produces the mangled name
path_to_spectrum_file = os.path.join(path_to_spectrum_files, reversed_mapping[filename])
#Try to resolve the full path
for key in mangled_mapping:
if displaying_filename in mangled_mapping[key]:
displaying_filename = mangled_mapping[key]
break
#loading file
spectrum_list = []
try:
spectrum_list = ming_spectrum_library.load_mzxml_file(path_to_spectrum_file)
except KeyboardInterrupt:
raise
except Exception as e:
print(e)
print("Could not load", path_to_spectrum_file)
spectrum_list = []
#structure_object is candidate from tsv, spectrum is the input features
#print(structures_by_filename)
for structure_object in structures_by_filename[filename]:
highest_intensity = -1000
best_spectrum = None
ppm_threshold = candidate_object["ppm_threshold"]
#print(structure_object)
#print("molecule mass","monoisotopic mass","ppm","filename","exact_mass","adduct")
for spectrum in spectrum_list:
if spectrum.ms_level == 1:
continue
#evaluate candidate_object
monoisotopic_mass = structure_object["monoisotopic_mass"]
#print(spectrum.mz, structure_object["exact_mass"])
mz_delta = abs(spectrum.mz - monoisotopic_mass)
ppm_delta = (mz_delta / monoisotopic_mass ) * 1000000
if ppm_delta > ppm_threshold:
continue
else:
if spectrum.get_total_spectrum_intensity() > highest_intensity:
#print(ppm_delta,ppm_threshold)
best_spectrum = spectrum
highest_intensity = max(spectrum.totIonCurrent, spectrum.get_total_spectrum_intensity())
if best_spectrum != None and highest_intensity > candidate_object["min_precursor_int"]:
#print(structure_object["monoisotopic_mass"],candidate_object["monoisotopic_mass"])
#output_dict["FILENAME"].append(os.path.basename(filename))
output_dict["FILENAME"].append(displaying_filename)
output_dict["SEQ"].append("*.*")
output_dict["COMPOUND_NAME"].append(structure_object["name"])
output_dict["MOLECULEMASS"].append(structure_object["monoisotopic_mass"])
output_dict["INSTRUMENT"].append(structure_object["instrument"])
output_dict["IONSOURCE"].append(structure_object["ionsource"])
output_dict["EXTRACTSCAN"].append(best_spectrum.scan)
output_dict["SMILES"].append(structure_object["smiles"])
output_dict["INCHI"].append(structure_object["inchi"])
output_dict["INCHIAUX"].append("N/A")
output_dict["CHARGE"].append(structure_object["charge"])
output_dict["IONMODE"].append(structure_object["ionmode"])
output_dict["PUBMED"].append(structure_object["pubmed"])
output_dict["ACQUISITION"].append(structure_object["acquisition"])
output_dict["EXACTMASS"].append(structure_object["exact_mass"])
output_dict["DATACOLLECTOR"].append(structure_object["datacollector"])
output_dict["ADDUCT"].append(structure_object["adduct"])
output_dict["INTEREST"].append("N/A")
output_dict["LIBQUALITY"].append("1")
output_dict["GENUS"].append("N/A")
output_dict["SPECIES"].append("N/A")
output_dict["STRAIN"].append("N/A")
output_dict["CASNUMBER"].append(structure_object["casnumber"])
output_dict["PI"].append(structure_object["pi"])
#print(spectrum.mz, monoisotopic_mass, ppm_delta, filename, structure_object["exact_mass"],structure_object["adduct"])
#print("Found ", structure_object["name"],structure_object["adduct"], highest_intensity)
#else:
#print("Not Seen", structure_object["name"], structure_object["adduct"], highest_intensity)
return output_dict
def main():
parser = argparse.ArgumentParser(description='Create parallel parameters')
parser.add_argument('toolname', help='name of input tool')
parser.add_argument('quantification_table', help='quantification_table')
parser.add_argument('quantification_table_reformatted',
help='quantification_table_reformatted')
parser.add_argument('input_spectra_folder', help='input_spectra_folder')
parser.add_argument('output_mgf', help='output_mgf')
parser.add_argument('workflowParameters', help='workflowParameters')
args = parser.parse_args()
input_filenames = glob.glob(os.path.join(args.input_spectra_folder, "*"))
if args.toolname == "MZMINE2":
print("MZMINE2")
if len(input_filenames) != 1:
print("Must input exactly 1 spectrum mgf file")
exit(1)
input_mgf = input_filenames[0]
shutil.copyfile(input_mgf, args.output_mgf)
mzmine2_formatter.convert_to_feature_csv(
args.quantification_table, args.quantification_table_reformatted)
elif args.toolname == "OPENMS":
print("OPENMS")
if len(input_filenames) != 1:
print("Must input exactly 1 spectrum mgf file")
exit(1)
input_mgf = input_filenames[0]
shutil.copyfile(input_mgf, args.output_mgf)
openms_formatter.convert_to_feature_csv(
args.quantification_table, args.quantification_table_reformatted)
elif args.toolname == "OPTIMUS":
print("OPTIMUS")
if len(input_filenames) != 1:
print("Must input exactly 1 spectrum mgf file")
exit(1)
input_mgf = input_filenames[0]
shutil.copyfile(input_mgf, args.output_mgf)
optimus_formatter.convert_to_feature_csv(
args.quantification_table, args.quantification_table_reformatted)
elif args.toolname == "MSDIAL":
print("MSDIAL")
if len(input_filenames) != 1:
print("Must input exactly 1 spectrum mgf file")
exit(1)
input_mgf = input_filenames[0]
shutil.copyfile(input_mgf, args.output_mgf)
msdial_formatter.convert_to_feature_csv(
args.quantification_table, args.quantification_table_reformatted)
elif args.toolname == "METABOSCAPE":
print("METABOSCAPE")
if len(input_filenames) != 1:
print("Must input exactly 1 spectrum mgf file")
exit(1)
input_mgf = input_filenames[0]
shutil.copyfile(input_mgf, args.output_mgf)
metaboscape_formatter.convert_to_feature_csv(
args.quantification_table, args.quantification_table_reformatted)
elif args.toolname == "XCMS3":
print("XCMS3")
if len(input_filenames) != 1:
print("Must input exactly 1 spectrum mgf file")
exit(1)
input_mgf = input_filenames[0]
shutil.copyfile(input_mgf, args.output_mgf)
xcms_formatter.convert_to_feature_csv(
args.quantification_table, args.quantification_table_reformatted)
elif args.toolname == "PROGENESIS":
print("PROGENESIS")
if len(input_filenames) != 1:
print("Must input exactly 1 spectrum mgf file")
exit(1)
input_mgf = input_filenames[0]
compound_scan_mapping = progenesis_formatter.convert_to_feature_csv(
args.quantification_table, args.quantification_table_reformatted)
progenesis_formatter.convert_mgf(input_mgf, args.output_mgf,
compound_scan_mapping)
elif args.toolname == "MZTABM":
print("MZTABM")
workflow_parameters = proteosafe.parse_xml_file(
args.workflowParameters)
mangled_mapping = proteosafe.get_mangled_file_mapping(
workflow_parameters)
name_mangle_mapping = {}
for key in mangled_mapping:
demangled_name = mangled_mapping[key]
name_mangle_mapping[os.path.basename(
demangled_name)] = os.path.join(args.input_spectra_folder, key)
compound_filename_mapping = mztabm_formatter.convert_to_feature_csv(
args.quantification_table, args.quantification_table_reformatted)
mztabm_formatter.create_mgf(input_filenames,
args.output_mgf,
compound_filename_mapping,
name_mangle_mapping=name_mangle_mapping)
def main():
parser = argparse.ArgumentParser(description='Create parallel parameters')
parser.add_argument('toolname', help='name of input tool')
parser.add_argument('quantification_table', help='quantification_table')
parser.add_argument('quantification_table_reformatted', help='quantification_table_reformatted')
parser.add_argument('input_spectra_folder', help='input_spectra_folder')
parser.add_argument('output_mgf', help='output_mgf')
parser.add_argument('workflowParameters', help='workflowParameters')
parser.add_argument('--QUANT_FILE_NORM', default="None", help='QUANT_FILE_NORM')
args = parser.parse_args()
input_filenames = glob.glob(os.path.join(args.input_spectra_folder, "*"))
if args.toolname == "MZMINE2":
print("MZMINE2")
if len(input_filenames) != 1:
print("Must input exactly 1 spectrum mgf file")
exit(1)
input_mgf = input_filenames[0]
shutil.copyfile(input_mgf, args.output_mgf)
mzmine2_formatter.convert_to_feature_csv(args.quantification_table, args.quantification_table_reformatted)
elif args.toolname == "OPENMS":
print("OPENMS")
if len(input_filenames) != 1:
print("Must input exactly 1 spectrum mgf file")
exit(1)
input_mgf = input_filenames[0]
shutil.copyfile(input_mgf, args.output_mgf)
openms_formatter.convert_to_feature_csv(args.quantification_table, args.quantification_table_reformatted)
elif args.toolname == "OPTIMUS":
print("OPTIMUS")
if len(input_filenames) != 1:
print("Must input exactly 1 spectrum mgf file")
exit(1)
input_mgf = input_filenames[0]
shutil.copyfile(input_mgf, args.output_mgf)
optimus_formatter.convert_to_feature_csv(args.quantification_table, args.quantification_table_reformatted)
elif args.toolname == "MSDIAL":
print("MSDIAL")
if len(input_filenames) != 1:
print("Must input exactly 1 spectrum mgf file")
exit(1)
input_mgf = input_filenames[0]
shutil.copyfile(input_mgf, args.output_mgf)
msdial_formatter.convert_to_feature_csv(args.quantification_table, args.quantification_table_reformatted)
elif args.toolname == "METABOSCAPE":
print("METABOSCAPE")
if len(input_filenames) != 1:
print("Must input exactly 1 spectrum mgf file")
exit(1)
input_mgf = input_filenames[0]
shutil.copyfile(input_mgf, args.output_mgf)
metaboscape_formatter.convert_to_feature_csv(args.quantification_table, args.quantification_table_reformatted)
elif args.toolname == "XCMS3":
print("XCMS3")
if len(input_filenames) != 1:
print("Must input exactly 1 spectrum mgf file")
exit(1)
input_mgf = input_filenames[0]
shutil.copyfile(input_mgf, args.output_mgf)
xcms_formatter.convert_to_feature_csv(args.quantification_table, args.quantification_table_reformatted)
elif args.toolname == "PROGENESIS":
print("PROGENESIS")
if len(input_filenames) != 1:
print("Must input exactly 1 spectrum mgf file")
exit(1)
input_mgf = input_filenames[0]
compound_scan_mapping = progenesis_formatter.convert_to_feature_csv(args.quantification_table, args.quantification_table_reformatted)
progenesis_formatter.convert_mgf(input_mgf, args.output_mgf, compound_scan_mapping)
elif args.toolname == "MZTABM":
print("MZTABM")
workflow_parameters = proteosafe.parse_xml_file(args.workflowParameters)
mangled_mapping = proteosafe.get_mangled_file_mapping(workflow_parameters)
name_mangle_mapping = {}
for key in mangled_mapping:
demangled_name = mangled_mapping[key]
name_mangle_mapping[os.path.basename(demangled_name)] = os.path.join(args.input_spectra_folder, key)
compound_filename_mapping = mztabm_formatter.convert_to_feature_csv(args.quantification_table, args.quantification_table_reformatted)
mztabm_formatter.create_mgf(input_filenames, args.output_mgf, compound_filename_mapping, name_mangle_mapping=name_mangle_mapping)
# Finally, we can renormlize the output
try:
if args.QUANT_FILE_NORM == "RowSum":
import pandas as pd
quant_df = pd.read_csv(args.quantification_table_reformatted, sep=",")
quant_df = quant_df.loc[:, ~quant_df.columns.str.contains('^Unnamed')]
for column in quant_df:
if "Peak area" in column:
quant_df[column] = quant_df[column] / sum(quant_df[column]) * 1000000
quant_df.to_csv(args.quantification_table_reformatted, sep=",", index=False)
except:
pass