def Agilent_reader(file_name):
if not isinstance(file_name, (str, pathlib.Path)):
raise TypeError(
"'file_name' must be a string or a pathlib.Path object")
if not isinstance(file_name, pathlib.Path):
file_name = pathlib.Path(file_name)
assert os.path.exists(file_name)
if (file_name / 'DATA.MS').exists():
d_file = open(file_name / 'DATA.MS', 'rb')
elif (file_name / 'data.ms').exists():
d_file = open(file_name / 'data.ms', 'rb')
else:
print(f'Error: {file_name} does not contain a data.ms file.')
raise ValueError(
f'Error: {file_name} does not contain a data.ms file.')
data = AgilentGCMSData()
options = Options()
load_file_info(d_file, data, options)
load_tic(d_file, data, options)
load_xic(d_file, data, options)
time_list = list(data.time)
rows, _ = data.xic.shape
scan_list = [Scan(data.mz, list(data.xic[r, :])) for r in range(rows)]
return GCMS_data(time_list, scan_list)
def ANDI_reader(file_name):
"""
A reader for ANDI-MS NetCDF files
:param file_name: The path of the ANDI-MS file
:type file_name: str or os.PathLike
:return: GC-MS data object
:rtype: :class:`pyms.GCMS.Class.GCMS_data`
:author: Qiao Wang
:author: Andrew Isaac
:author: Vladimir Likic
:author: Dominic Davis-Foster
"""
if not isinstance(file_name, (str, pathlib.Path)):
raise TypeError(
"'file_name' must be a string or a pathlib.Path object")
rootgrp = Dataset(file_name, "r+", format='NETCDF3_CLASSIC')
# TODO: find out if netCDF4 throws specific errors that we can use here
print(f" -> Reading netCDF file '{file_name}'")
scan_list = []
mass = rootgrp.variables[__MASS_STRING][:]
intensity = rootgrp.variables[__INTENSITY_STRING][:]
scan_lengths = rootgrp.variables[
"point_count"] # The number of data points in each scan
mass_values = mass.tolist()
intensity_values = intensity.tolist()
if len(mass_values) != len(intensity_values):
raise ValueError("The lengths of the mass and intensity lists differ!")
offset = 0
for idx, length in enumerate(scan_lengths):
mass_list = mass_values[offset:offset + length]
assert len(mass_values[offset:offset + length]) == length
intensity_list = intensity_values[offset:offset + length]
assert len(intensity_values[offset:offset + length]) == length
scan_list.append(Scan(mass_list, intensity_list))
offset += length
assert offset == len(mass_values)
time = rootgrp.variables[__TIME_STRING][:]
time_list = time.tolist()
# sanity check
if not len(time_list) == len(scan_list):
raise ValueError(
"number of time points does not equal the number of scans")
return GCMS_data(time_list, scan_list)
def agilent_reader(
file_name: PathLike) -> GCMS_data: # pragma: no cover (!Windows)
"""
Reader for Agilent MassHunter ``.d`` files.
:param file_name: Path of the file to read.
:return: GC-MS data object.
"""
if not isinstance(file_name, (str, pathlib.Path)):
raise TypeError(
"'file_name' must be a string or a pathlib.Path object")
if not isinstance(file_name, pathlib.Path):
file_name = pathlib.Path(file_name)
print(f" -> Reading Agilent data file '{file_name}'")
time_list = []
scan_list = []
reader = MassSpecDataReader(file_name)
for scan_no in range(
reader.file_information.ms_scan_file_info.total_scans):
spectrum = reader.get_spectrum_by_scan(scan_no)
scan_list.append(Scan(spectrum.x_data, spectrum.y_data))
time_list.append(mean(spectrum.acquired_time_ranges[0]) * 60.0)
# sanity check
time_len = len(time_list)
scan_len = len(scan_list)
if not time_len == scan_len: # pragma: no cover
print(time_list)
print(scan_list)
raise ValueError(
f"Number of time points ({time_len}) does not equal the number of scans ({scan_len})."
)
data = GCMS_data(time_list, scan_list)
return data
def JCAMP_reader(file_name: Union[str, os.PathLike]) -> GCMS_data:
"""
Generic reader for JCAMP DX files
:param file_name: Path of the file to read
:type file_name: str or os.PathLike
:return: GC-MS data object
:rtype: :class:`pyms.GCMS.Class.GCMS_data`
:authors: Qiao Wang, Andrew Isaac, Vladimir Likic, David Kainer, Dominic Davis-Foster (pathlib support)
"""
if not is_path(file_name):
raise TypeError("'file_name' must be a string or a PathLike object")
file_name = prepare_filepath(file_name, mkdirs=False)
print(f" -> Reading JCAMP file '{file_name}'")
lines_list = file_name.open('r')
data = []
page_idx = 0
xydata_idx = 0
time_list = []
scan_list = []
header_info = {} # Dictionary containing header information
for line in lines_list:
if len(line.strip()) != 0:
# prefix = line.find('#')
# if prefix == 0:
if line.startswith("##"):
# key word or information
fields = line.split('=', 1)
fields[0] = fields[0].lstrip("##").upper()
fields[1] = fields[1].strip()
if "PAGE" in fields[0]:
if "T=" in fields[1]:
# PAGE contains retention time starting with T=
# FileConverter Pro style
time = float(fields[1].lstrip(
"T=")) # rt for the scan to be submitted
time_list.append(time)
page_idx = page_idx + 1
elif "RETENTION_TIME" in fields[0]:
# OpenChrom style
time = float(fields[1]) # rt for the scan to be submitted
# Check to make sure time is not already in the time list;
# Can happen when both ##PAGE and ##RETENTION_TIME are specified
if time_list[-1] != time:
time_list.append(time)
elif fields[0] in xydata_tags:
xydata_idx = xydata_idx + 1
elif fields[0] in header_info_fields:
if fields[1].isdigit():
header_info[fields[0]] = int(fields[1])
elif is_float(fields[1]):
header_info[fields[0]] = float(fields[1])
else:
header_info[fields[0]] = fields[1]
# elif prefix == -1:
else:
# Line doesn't start with ##
# data
if page_idx > 1 or xydata_idx > 1:
if len(data) % 2 == 1:
# TODO: This means the data is not in x, y pairs
# Make a better error message
raise ValueError("data not in pair !")
mass_list = []
intensity_list = []
for i in range(len(data) // 2):
mass_list.append(data[i * 2])
intensity_list.append(data[i * 2 + 1])
if len(mass_list) != len(intensity_list):
raise ValueError(
"len(mass_list) is not equal to len(intensity_list)"
)
scan_list.append(Scan(mass_list, intensity_list))
data = []
data_sub = line.strip().split(',')
for item in data_sub:
if not len(item.strip()) == 0:
data.append(float(item.strip()))
if page_idx > 1:
page_idx = 1
if xydata_idx > 1:
xydata_idx = 1
else:
data_sub = line.strip().split(',')
for item in data_sub:
if not len(item.strip()) == 0:
data.append(float(item.strip()))
if len(data) % 2 == 1:
# TODO: This means the data is not in x, y pairs
# Make a better error message
raise ValueError("data not in pair !")
# get last scan
mass = []
intensity = []
for i in range(len(data) // 2):
mass.append(data[i * 2])
intensity.append(data[i * 2 + 1])
if len(mass) != len(intensity):
raise ValueError("len(mass) is not equal to len(intensity)")
scan_list.append(Scan(mass, intensity))
# sanity check
time_len = len(time_list)
scan_len = len(scan_list)
if time_len != scan_len:
print(time_list)
print(scan_list)
raise ValueError(
f"Number of time points ({time_len}) does not equal the number of scans ({scan_len})"
)
data = GCMS_data(time_list, scan_list)
return data
def mzML_reader(file_name):
"""
A reader for mzML files
:param file_name: The name of the mzML file
:type file_name: str or pathlib.Path
:return: GC-MS data object
:rtype: :class:`pyms.GCMS.Class.GCMS_data`
:author: Sean O'Callaghan
:author: Dominic Davis-Foster (pathlib support)
"""
if not is_path(file_name):
raise TypeError("'file_name' must be a string or a PathLike object")
mzml_file = pymzml.run.Reader(str(file_name))
try: # avoid printing from each rank
comm = MPI.COMM_WORLD
rank = comm.Get_rank()
size = comm.Get_size()
if rank == 0:
file_names = []
for i in range(1, size):
recv_buffer = ""
file_n = comm.recv(recv_buffer, i)
file_names.append(file_n)
print(" -> Reading mzML files:")
print(file_name)
for file_n in file_names:
print(file_n)
else:
comm.send(file_name, dest=0)
# TODO: Find specific error
except Exception as e:
print(e)
print(f" -> Reading mzML file '{file_name}'")
scan_list = []
time_list = []
for spectrum in mzml_file:
mass_list = []
intensity_list = []
for mz, i in spectrum.peaks:
mass_list.append(mz)
intensity_list.append(i)
# scan_list.append(Scan(mass_list, intensity_list))
for element in spectrum.xmlTree:
# For some reason there are spectra with no time value,
# Ignore these????????????
if element.get('accession') == "MS:1000016": # time value
# We need time in seconds not minutes
time_list.append(60 * float(element.get('value')))
scan_list.append(Scan(mass_list, intensity_list))
# print("time:", len(time_list))
# print("scan:", len(scan_list))
data = GCMS_data(time_list, scan_list)
return data
def test_zero_length():
# TODO: finish
scan = Scan([], [])
def test_equality(im, scan):
assert scan == Scan(scan.mass_list, scan.intensity_list)
assert scan != im.get_scan_at_index(1234)
def test_errors(scan, obj, expects):
with pytest.raises(expects):
Scan(obj, scan.intensity_list)
with pytest.raises(expects):
Scan(scan.mass_list, obj)
