def write_intensities_stream(self, file_name: PathLike):
"""
Loop over all scans and, for each scan, write the intensities to the
given file, one intensity per line.
Intensities from different scans are joined without any delimiters.
:param file_name: Output file name.
:authors: Vladimir Likic, Dominic Davis-Foster (pathlib support)
""" # noqa: D400
if not is_path(file_name):
raise TypeError(
"'file_name' must be a string or a PathLike object")
file_name = prepare_filepath(file_name)
# n = len(self._scan_list)
print(" -> Writing scans to a file")
fp = file_name.open('w', encoding="UTF-8")
for scan in self._scan_list:
intensities = scan.intensity_list
for i in intensities:
fp.write(f"{i:8.4f}\n")
fp.close()
def run(self, filename: PathLike):
"""
Parse configuration from the given file.
:param filename: The filename of the YAML configuration file.
"""
filename = PathPlus(filename)
if not filename.is_file():
raise FileNotFoundError(str(filename))
with tempfile.TemporaryDirectory() as tmpdir:
tmpdir_p = PathPlus(tmpdir)
schema_file = tmpdir_p / "schema.json"
schema = make_schema(*self.config_vars)
schema["additionalProperties"] = self.allow_unknown_keys
schema_file.dump_json(schema)
validate_files(schema_file, filename)
parsed_config_vars: MutableMapping[str, Any] = {}
with filename.open() as file:
raw_config_vars: Mapping[str, Any] = YAML(typ="safe",
pure=True).load(file)
for var in self.config_vars:
parsed_config_vars[var.__name__] = getattr(
self, f"visit_{var.__name__}", var.get)(raw_config_vars)
return self.custom_parsing(raw_config_vars, parsed_config_vars,
filename)
def read_expr_list(file_name: PathLike) -> List[Experiment]:
"""
Reads the set of experiment files and returns a list of :class:`pyms.Experiment.Experiment` objects.
:param file_name: The name of the file which lists experiment dump file names, one file per line.
:return: A list of Experiment instances.
:author: Vladimir Likic
"""
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)
fp = file_name.open(encoding="UTF-8")
exprfiles = fp.readlines()
fp.close()
expr_list = []
for exprfile in exprfiles:
exprfile = exprfile.strip()
expr = load_expr(exprfile)
expr_list.append(expr)
return expr_list
def write(self,
file_name: PathLike,
minutes: bool = False,
formatting: bool = True):
"""
Writes the ion chromatogram to the specified file.
:param file_name: The name of the output file
:param minutes: A boolean value indicating whether to write time in minutes
:param formatting: Whether to format the numbers in the output.
:authors: Lewis Lee, Vladimir Likic, 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)
with file_name.open('w', encoding="UTF-8") as fp:
time_list = copy.deepcopy(self._time_list)
if minutes:
for ii in range(len(time_list)):
time_list[ii] = time_list[ii] / 60.0
for ii in range(len(time_list)):
if formatting:
fp.write(
f"{time_list[ii]:8.4f} {self._intensity_array[ii]:#.6e}\n"
)
else:
fp.write(f"{time_list[ii]} {self._intensity_array[ii]}\n")
def get_metadata_for_file(filename: PathLike) -> Dict[str, Any]:
"""
Returns the EXIF metadata for ``filename``, as a ``key: value`` mapping.
:param filename:
"""
filename = PathPlus(filename)
if not filename.is_file():
raise FileNotFoundError(filename)
# get the tags
with filename.open("rb") as fp:
data = exifread.process_file(fp, details=False, debug=False)
if data:
return {k: str(v) for k, v in data.items()}
else:
# using exiftool as a backup for some files including videos
with exiftool.ExifTool() as et:
try:
data = et.get_metadata(str(filename))
except json.decoder.JSONDecodeError:
raise ValueError(
f"Could not parse EXIF data for {filename} or no EXIF data found."
)
return dict(data)
def from_jcamp(cls: Type[_M], file_name: PathLike) -> _M:
"""
Create a MassSpectrum from a JCAMP-DX file.
:param file_name: Path of the file to read.
:authors: Qiao Wang, Andrew Isaac, Vladimir Likic, David Kainer, Dominic Davis-Foster
"""
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', encoding="UTF-8")
xydata = []
last_tag = None
for line in lines_list:
if line.strip():
if line.startswith("##"):
# key word or information
fields = line.split('=', 1)
current_tag = fields[0] = fields[0].lstrip("##").upper()
last_tag = fields[0]
if current_tag.upper().startswith("END"):
break
else:
if last_tag in xydata_tags:
line_sub = re.split(r",| ", line.strip())
for item in line_sub:
if not len(item.strip()) == 0:
xydata.append(float(item.strip()))
# By this point we should have all of the xydata
if len(xydata) % 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(xydata) // 2):
mass_list.append(xydata[i * 2])
intensity_list.append(xydata[i * 2 + 1])
return cls(mass_list, intensity_list)
def export_leco_csv(self, file_name: PathLike):
"""
Exports data in LECO CSV format.
:param file_name: The name of the output file.
:authors: Andrew Isaac, Vladimir Likic, 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)
if not file_name.parent.is_dir():
file_name.parent.mkdir(parents=True)
mass_list = self._mass_list
time_list = self._time_list
vals = self._intensity_array
fp = file_name.open('w', encoding="UTF-8")
# Format is text header with:
# "Scan","Time",...
# and the rest is "TIC" or m/z as text, i.e. "50","51"...
# The following lines are:
# scan_number,time,value,value,...
# scan_number is an int, rest seem to be fixed format floats.
# The format is 0.000000e+000
# write header
fp.write('"Scan","Time"')
for ii in mass_list:
if is_number(ii):
fp.write(f',"{int(ii):d}"')
else:
raise TypeError("mass list datum not a number")
fp.write("\r\n") # windows CR/LF
# write lines
for ii, time_ in enumerate(time_list):
fp.write(f"{ii},{time_:#.6e}")
for jj in range(len(vals[ii])):
if is_number(vals[ii][jj]):
fp.write(f",{vals[ii][jj]:#.6e}")
else:
raise TypeError("datum not a number")
fp.write("\r\n")
fp.close()
def load_object(file_name: PathLike) -> object:
"""
Loads an object previously dumped with :func:`~.dump_object`.
:param file_name: Name of the object dump file.
:return: Object contained in the file.
:authors: Vladimir Likic, 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)
with file_name.open("wb") as fp:
return pickle.load(fp)
def file_lines(file_name: PathLike, strip: bool = False) -> List[str]:
"""
Returns lines from a file, as a list.
:param file_name: Name of a file
:param strip: If True, lines are pre-processed. Newline characters are
removed, leading and trailing whitespaces are removed, and lines
starting with '#' are discarded
:return: A list of lines
:authors: Vladimir Likic, 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)
with file_name.open(encoding="UTF-8") as fp:
lines = fp.readlines()
if strip:
# strip leading and trailing whitespaces
lines_filtered = []
for line in lines:
line = line.strip()
lines_filtered.append(line)
# discard comments
lines_to_discard = []
for line in lines_filtered:
# remove empty lines and comments
if len(line) == 0 or line[0] == '#':
lines_to_discard.append(line)
for line in lines_to_discard:
lines_filtered.remove(line)
lines = lines_filtered
return lines
def dump_to_file(self,
data: Union[MutableMapping, Sequence],
filename: PathLike,
mode: str = 'w'):
"""
Dump the given data to the specified file.
:param data:
:param filename:
:param mode:
"""
filename = PathPlus(filename)
if 'w' in mode:
filename.write_lines([
"# Configuration for 'repo_helper' (https://github.com/repo-helper/repo_helper)",
self.dumps(data, explicit_start=True),
])
elif 'a' in mode:
with filename.open('a') as fp:
fp.write('\n')
fp.write(self.dumps(data, explicit_start=False))
def import_leco_csv(file_name: PathLike) -> IntensityMatrix:
"""
Imports data in LECO CSV format.
:param file_name: Path of the file to read.
:return: Data as an IntensityMatrix.
:authors: Andrew Isaac, 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)
lines_list = file_name.open('r')
data = []
time_list = []
mass_list = []
# Format is text header with:
# "Scan","Time",...
# and the rest is "TIC" or m/z as text, i.e. "50","51"...
# The following lines are:
# scan_number,time,value,value,...
# scan_number is an int, rest seem to be fixed format floats.
# The format is 0.000000e+000
num_mass = 0
FIRST = True
HEADER = True
data_col = -1
time_col = -1
# get each line
for line in lines_list:
cols = -1
data_row = []
if len(line.strip()) > 0:
data_list = line.strip().split(',')
# get each value in line
for item in data_list:
item = item.strip()
item = item.strip("'\"") # remove quotes (in header)
# Get header
if HEADER:
cols += 1
if len(item) > 0:
if item.lower().find("time") > -1:
time_col = cols
try:
value = float(item)
# find 1st col with number as header
if FIRST and value > 1: # assume >1 mass
data_col = cols
# assume time col is previous col
if time_col < 0:
time_col = cols - 1
FIRST = False
mass_list.append(value)
num_mass += 1
except ValueError:
pass
# Get rest
else:
cols += 1
if len(item) > 0:
try:
value = float(item)
if cols == time_col:
time_list.append(value)
elif cols >= data_col:
data_row.append(value)
except ValueError:
pass
# check row length
if not HEADER:
if len(data_row) == num_mass:
data.append(data_row)
else:
warn("ignoring row")
HEADER = False
# check col lengths
if len(time_list) != len(data):
warn("number of data rows and time list length differ")
return IntensityMatrix(time_list, mass_list, data)
def write_mass_hunter_csv(
alignment: Alignment,
file_name: PathLike,
top_ion_list: List[int],
): # , peak_list_name):
"""
Creates a csv file with UID, common and qualifying ions and their
ratios for mass hunter interpretation.
:param alignment: alignment object to write to file
:param file_name: name of the output file.
:param top_ion_list: a list of the common ions for each peak in the
averaged peak list for the alignment.
""" # noqa: D400
if not is_path(file_name):
raise TypeError("'file_name' must be a string or a PathLike object")
file_name = prepare_filepath(file_name)
fp = file_name.open('w', encoding="UTF-8")
if top_ion_list is None:
raise ValueError("List of common ions must be supplied")
# write headers
fp.write(
'"UID","Common Ion","Qual Ion 1","ratio QI1/CI","Qual Ion 2",'
'"ratio QI2/CI","l window delta","r window delta"\n'
)
rtsums: List[float] = []
rtcounts = []
# The following two arrays will become list of lists
# such that:
# areas = [ [align1_peak1, align2_peak1, .....,alignn_peak1]
# [align1_peak2, ................................]
# .............................................
# [align1_peakm,....................,alignn_peakm] ]
areas = [] # type: ignore
new_peak_lists = [] # type: ignore
rtmax = []
rtmin = []
for peak_list in alignment.peakpos:
index = 0
for peak in peak_list:
# on the first iteration, populate the lists
if len(areas) < len(peak_list):
areas.append([])
new_peak_lists.append([])
rtsums.append(0)
rtcounts.append(0)
rtmax.append(0.0)
rtmin.append(0.0)
if peak is not None:
rt = peak.rt
# get the area of the common ion for the peak
# an area of 'na' shows that while the peak was
# aligned, the common ion was not present
area = peak.get_ion_area(top_ion_list[index])
areas[index].append(area)
new_peak_lists[index].append(peak)
# The following code to the else statement is
# just for calculating the average rt
rtsums[index] += rt
rtcounts[index] += 1
# quick workaround for weird problem when
# attempting to set rtmin to max time above
if rtmin[index] == 0.0:
rtmin[index] = 5400.0
if rt > rtmax[index]:
rtmax[index] = rt
if rt < rtmin[index]:
rtmin[index] = rt
else:
areas[index].append(None)
index += 1
out_strings = []
compo_peaks = []
index = 0
# now write the strings for the file
for area_list in areas:
# write initial info:
# peak unique id, peak average rt
compo_peak = composite_peak(new_peak_lists[index])
if compo_peak is None:
continue
compo_peaks.append(compo_peak)
peak_UID = compo_peak.UID
peak_UID_string = f'"{peak_UID}"'
# calculate the time from the leftmost peak to the average
l_window_delta = compo_peak.rt - rtmin[index]
# print("l_window", l_window_delta, "rt", compo_peak.rt, "rt_min", rtmin[index])
r_window_delta = rtmax[index] - compo_peak.rt
common_ion = top_ion_list[index]
qual_ion_1 = int(peak_UID_string.split('-')[0].strip('"'))
qual_ion_2 = int(peak_UID_string.split('-')[1])
if qual_ion_1 == common_ion:
qual_ion_1 = compo_peak.get_third_highest_mz()
elif qual_ion_2 == common_ion:
qual_ion_2 = compo_peak.get_third_highest_mz()
else:
pass
ci_intensity = compo_peak.get_int_of_ion(common_ion)
q1_intensity = compo_peak.get_int_of_ion(qual_ion_1)
q2_intensity = compo_peak.get_int_of_ion(qual_ion_2)
try:
q1_ci_ratio = float(q1_intensity) / float(ci_intensity)
except TypeError: # if no area available for that ion
q1_ci_ratio = 0.0
except ZeroDivisionError:
# shouldn't happen but does!!
q1_ci_ratio = 0.01
try:
q2_ci_ratio = float(q2_intensity) / float(ci_intensity)
except TypeError:
q2_ci_ratio = 0.0
except ZeroDivisionError:
# shouldn't happen, but does!!
q2_ci_ratio = 0.01
out_strings.append(
','.join([
peak_UID,
f"{common_ion}",
f"{qual_ion_1}",
f"{q1_ci_ratio * 100:.1f}",
f"{qual_ion_2}",
f"{q2_ci_ratio * 100:.1f}",
f"{(l_window_delta + 1.5) / 60:.2f}",
f"{(r_window_delta + 1.5) / 60:.2f}",
])
)
index += 1
# now write the file
# print("length of areas[0]", len(areas[0]))
# print("lenght of areas", len(areas))
# print("length of out_strings", len(out_strings))
for row in out_strings:
fp.write(f"{row}\n")
# dump_object(compo_peaks, peak_list_name)
fp.close()
