def peaklist(self, scan_id, function_noise="median"):
if function_noise not in ["mean", "median", "mad"]:
raise ValueError("select a function that is available [mean, median, mad]")
for scan in self.run():
if scan["id"] == scan_id:
mzs, ints = zip(*scan.peaks)
scan_time = scan["MS:1000016"]
tic = scan["total ion current"]
if "MS:1000927" in scan:
ion_injection_time = scan["MS:1000927"]
else:
ion_injection_time = None
header = scan['MS:1000512']
mz_range = mz_range_from_header(header)
ms_level = scan['ms level']
pl = PeakList(ID=scan["id"], mz=mzs, intensity=ints,
mz_range=mz_range,
header=header,
ms_level=ms_level,
ion_injection_time=ion_injection_time,
scan_time=scan_time,
tic=tic,
function_noise=function_noise)
snr = np.divide(ints, scan.estimatedNoiseLevel(mode=function_noise))
pl.add_attribute('snr', snr)
return pl
return None
def remove_edges(pls_sd):
if type(pls_sd) is not dict and type(
pls_sd) is not collections.OrderedDict:
raise TypeError(
"Incorrect format - dict or collections.OrderedDict required")
mzrs = [mz_range_from_header(h) for h in pls_sd]
new_mzrs = _calculate_edges(mzrs)
for h in pls_sd.keys():
mz_ranges = len(pls_sd[h]) * [new_mzrs[pls_sd.keys().index(h)]]
for i in range(len(pls_sd[h])):
remove = [
np.where(pls_sd[h][i].mz == mz)[0][0] for mz in pls_sd[h][i].mz
if mz < mz_ranges[i][0] or mz >= mz_ranges[i][1]
]
for mz in pls_sd[h][i].mz:
if mz < mz_ranges[i][0] or mz >= mz_ranges[i][1]:
remove.extend(list(np.where(pls_sd[h][i].mz == mz)[0]))
pls_sd[h][i].remove_peak(remove)
return pls_sd
def peaklist(self, scan_id, function_noise="median"):
if function_noise not in ["mean", "median", "mad"]:
raise ValueError(
"select a function that is available [mean, median, mad]")
run = pymzml.run.Reader(self.filename)
for scan in run:
if scan["id"] == scan_id:
peaks = scan.peaks("raw")
if len(peaks) > 0:
mzs, ints = list(zip(*peaks))
else:
mzs, ints = [], []
scan_time = scan["MS:1000016"]
tic = scan["total ion current"]
if "MS:1000927" in scan:
ion_injection_time = scan["MS:1000927"]
else:
ion_injection_time = None
header = scan['MS:1000512']
mz_range = mz_range_from_header(header)
ms_level = scan['ms level']
pl = PeakList(ID=scan["id"],
mz=mzs,
intensity=ints,
mz_range=mz_range,
header=header,
ms_level=ms_level,
ion_injection_time=ion_injection_time,
scan_time=scan_time,
tic=tic,
function_noise=function_noise)
snr = np.divide(
ints, scan.estimated_noise_level(mode=function_noise))
pl.add_attribute('snr', snr)
run.info["file_object"].close()
return pl
return None
def read_scans(fn,
source,
function_noise,
min_scans=1,
filter_scan_events=None):
if filter_scan_events is None:
filter_scan_events = {}
if not fn.lower().endswith(".mzml") and not fn.lower().endswith(".raw"):
raise IOError("Check format raw data (.RAW or .mzML)")
if min_scans is not None and type(min_scans) is not int:
raise ValueError("Integer (>= 1) or None required for min_scans")
if zipfile.is_zipfile(source):
if fn.lower().endswith(".mzml"):
run = mzml_portal.Mzml(fn, source)
elif fn.lower().endswith(".raw"):
raise IOError("Zip file with raw files not supported")
else:
raise IOError("Incorrect format: {}".format(os.path.basename(fn)))
else:
if fn.lower().endswith(".mzml"):
run = mzml_portal.Mzml(fn)
elif fn.lower().endswith(".raw"):
run = thermo_raw_portal.ThermoRaw(fn)
else:
raise IOError("Incorrect format: {}".format(os.path.basename(fn)))
h_sids = run.headers()
if type(filter_scan_events) is dict and len(filter_scan_events) > 0:
if ("include" in filter_scan_events and "exclude" in filter_scan_events) or \
("include" not in filter_scan_events and "exclude" not in filter_scan_events):
raise ValueError(
"Use 'exclude' or 'include' for filter_scan_events not both. E.g {'include': [[70.0, 170.0, 'sim']]}"
)
if len([
True for fse in filter_scan_events.values()[0] if len(fse) == 3
]) != len(filter_scan_events.values()[0]):
raise ValueError(
"Provide a start, end and scan type (sim or full) for filter_scan_events."
)
filter_scan_events = {
filter_scan_events.keys()[0]:
[[float(fse[0]), float(fse[1]),
str(fse[2])] for fse in filter_scan_events.values()[0]]
}
h_descs = {}
for h in h_sids.copy():
mzr = mz_range_from_header(h)
h_descs[h] = [mzr[0], mzr[1], scan_type_from_header(h).lower()]
incl_excl = filter_scan_events.keys()[0]
for hd in filter_scan_events[incl_excl]:
if hd not in h_descs.values():
logging.warning("Event {} doest not exist".format(str(hd)))
for hd in h_descs:
if filter_scan_events.keys()[0] == "include":
if h_descs[hd] not in filter_scan_events["include"]:
del h_sids[hd]
elif filter_scan_events.keys()[0] == "exclude":
if h_descs[hd] in filter_scan_events["exclude"]:
del h_sids[hd]
if len(h_sids) == 0:
raise Exception("No scan data to process. Check filter_scan_events")
scans = collections.OrderedDict()
for h, sids in h_sids.iteritems():
if len(sids) >= min_scans:
scans[h] = run.peaklists(sids, function_noise)
else:
logging.warning(
'Not enough scans for [{}] [{} < {}]. Scan event {} has been removed.'
.format(h, len(scans), min_scans, h))
if fn.lower().endswith(".raw"):
run.close()
return scans