def missing_peak_finder(sample, filename, points=13, null_ions=[73, 147],\
crop_ions=[50,540], threshold=1000, rt_window=1, filetype='mzml'):
"""
@summary: Integrates raw data around missing peak locations
to fill in NAs in the data matrix
@param sample: The sample object containing missing peaks
@type sample: pyms.MissingPeak.Class.Sample
@param andi_file: Name of the raw data file
@type andi_file: stringType
@param points: Peak finding - Peak if maxima over 'points' \
number of scans (Default 3)
@type points: intType
@param null_ions: Ions to be deleted in the matrix
@type null_ions: listType
@param crop_ions: Range of Ions to be considered
@type crop_ions: listType
@param threshold: Minimum intensity of IonChromatogram allowable to fill\
missing peak
@type threshold: intType
@param rt_window: Window in seconds around average RT to look for \
missing peak
@type rt_window: floatType
@param filetype: either mzml or netcdf
@type filetype: stringType
@author: Sean O'Callaghan
"""
### some error checks on null and crop ions
### a for root,files,dirs in os.path.walk(): loop
print "Sample:", sample.get_name(), "File:", filename
if filetype.lower() == 'cdf':
data = ANDI_reader(filename)
elif filetype.lower() == 'mzml':
data = mzML_reader(filename)
else:
print "file type not valid"
# build integer intensity matrix
im = build_intensity_matrix_i(data)
for null_ion in null_ions:
im.null_mass(null_ion)
im.crop_mass(crop_ions[0], crop_ions[1])
# get the size of the intensity matrix
n_scan, n_mz = im.get_size()
# smooth data
for ii in range(n_mz):
ic = im.get_ic_at_index(ii)
ic1 = savitzky_golay(ic, points)
ic_smooth = savitzky_golay(ic1, points)
ic_base = tophat(ic_smooth, struct="1.5m")
im.set_ic_at_index(ii, ic_base)
for mp in sample.get_missing_peaks():
mp_rt = mp.get_rt()
common_ion = mp.get_ci()
qual_ion_1 = float(mp.get_qual_ion1())
qual_ion_2 = float(mp.get_qual_ion2())
ci_ion_chrom = im.get_ic_at_mass(common_ion)
print "ci = ", common_ion
qi1_ion_chrom = im.get_ic_at_mass(qual_ion_1)
print "qi1 = ", qual_ion_1
qi2_ion_chrom = im.get_ic_at_mass(qual_ion_2)
print "qi2 = ", qual_ion_2
######
# Integrate the CI around that particular RT
#######
#Convert time to points
# How long between scans?
points_1 = ci_ion_chrom.get_index_at_time(float(mp_rt))
points_2 = ci_ion_chrom.get_index_at_time(float(mp_rt) - rt_window)
print "rt_window = ", points_1 - points_2
rt_window_points = points_1 - points_2
maxima_list = get_maxima_list_reduced(ci_ion_chrom, mp_rt, \
rt_window_points)
large_peaks = []
for rt, intens in maxima_list:
if intens > threshold:
q1_index = qi1_ion_chrom.get_index_at_time(rt)
q2_index = qi2_ion_chrom.get_index_at_time(rt)
q1_intensity = qi1_ion_chrom.get_intensity_at_index(q1_index)
q2_intensity = qi2_ion_chrom.get_intensity_at_index(q2_index)
if q1_intensity > threshold / 2 and q2_intensity > threshold / 2:
large_peaks.append([rt, intens])
print('found %d peaks above threshold' % len(large_peaks))
areas = []
for peak in large_peaks:
apex = ci_ion_chrom.get_index_at_time(peak[0])
ia = ci_ion_chrom.get_intensity_array().tolist()
area, left, right, l_share, r_share = ion_area(ia, apex, 0)
areas.append(area)
########################
areas.sort()
if len(areas) > 0:
biggest_area = areas[-1]
mp.set_ci_area(biggest_area)
mp.set_exact_rt("{:.3f}".format(float(mp_rt) / 60.0))
print "found area:", biggest_area, "at rt:", mp_rt
else:
print "Missing peak at rt = ", mp_rt
mp.set_ci_area('na')
def missing_peak_finder(sample, filename, points=13, null_ions=[73, 147],\
crop_ions=[50,540], threshold=1000, rt_window=1, filetype='cdf'):
"""
@summary: Integrates raw data around missing peak locations
to fill in NAs in the data matrix
@param sample: The sample object containing missing peaks
@type sample: pyms.MissingPeak.Class.Sample
@param andi_file: Name of the raw data file
@type andi_file: stringType
@param points: Peak finding - Peak if maxima over 'points' \
number of scans (Default 3)
@type points: intType
@param null_ions: Ions to be deleted in the matrix
@type null_ions: listType
@param crop_ions: Range of Ions to be considered
@type crop_ions: listType
@param threshold: Minimum intensity of IonChromatogram allowable to fill\
missing peak
@type threshold: intType
@param rt_window: Window in seconds around average RT to look for \
missing peak
@type rt_window: floatType
@author: Sean O'Callaghan
"""
### some error checks on null and crop ions
### a for root,files,dirs in os.path.walk(): loop
print "Sample:", sample.get_name(), "File:", filename
if filetype == 'cdf':
data = ANDI_reader(filename)
elif filetype == 'mzml':
data = mzML_reader(filename)
else:
print "file type not valid"
# build integer intensity matrix
im = build_intensity_matrix_i(data)
for null_ion in null_ions:
im.null_mass(null_ion)
im.crop_mass(crop_ions[0], crop_ions[1])
# get the size of the intensity matrix
n_scan, n_mz = im.get_size()
# smooth data
for ii in range(n_mz):
ic = im.get_ic_at_index(ii)
ic1 = savitzky_golay(ic, points)
ic_smooth = savitzky_golay(ic1, points)
ic_base = tophat(ic_smooth, struct="1.5m")
im.set_ic_at_index(ii, ic_base)
for mp in sample.get_missing_peaks():
mp_rt = mp.get_rt()
common_ion = mp.get_ci()
qual_ion_1 = float(mp.get_qual_ion1())
qual_ion_2 = float(mp.get_qual_ion2())
ci_ion_chrom = im.get_ic_at_mass(common_ion)
print "ci = ",common_ion
qi1_ion_chrom = im.get_ic_at_mass(qual_ion_1)
print "qi1 = ", qual_ion_1
qi2_ion_chrom = im.get_ic_at_mass(qual_ion_2)
print "qi2 = ", qual_ion_2
######
# Integrate the CI around that particular RT
#######
#Convert time to points
# How long between scans?
points_1 = ci_ion_chrom.get_index_at_time(float(mp_rt))
points_2 = ci_ion_chrom.get_index_at_time(float(mp_rt)-rt_window)
print "rt_window = ", points_1 - points_2
rt_window_points = points_1 - points_2
maxima_list = get_maxima_list_reduced(ci_ion_chrom, mp_rt, \
rt_window_points)
large_peaks = []
for rt, intens in maxima_list:
if intens > threshold:
q1_index = qi1_ion_chrom.get_index_at_time(rt)
q2_index = qi2_ion_chrom.get_index_at_time(rt)
q1_intensity = qi1_ion_chrom.get_intensity_at_index(q1_index)
q2_intensity = qi2_ion_chrom.get_intensity_at_index(q2_index)
if q1_intensity > threshold/2 and q2_intensity > threshold/2:
large_peaks.append([rt, intens])
print('found %d peaks above threshold'%len(large_peaks))
areas = []
for peak in large_peaks:
apex = ci_ion_chrom.get_index_at_time(peak[0])
ia = ci_ion_chrom.get_intensity_array().tolist()
area, left, fight, l_share, r_share = ion_area(ia, apex, 0)
areas.append(area)
########################
areas.sort()
if len(areas)>0:
biggest_area = areas[-1]
mp.set_ci_area(biggest_area)
print "found area:", biggest_area, "at rt:", mp_rt
else:
print "Missing peak at rt = ", mp_rt
mp.set_ci_area('na')
"""proc.py
"""
import sys
sys.path.append("/x/PyMS")
from pyms.GCMS.IO.MZML.Function import mzML_reader
# read the raw data
mzml_file = "/x/PyMS/data/TP1U-11-16_86-2207.mzML"
data = mzML_reader(mzml_file)
# raw data operations
print "minimum mass found in all data: ", data.get_min_mass()
print "maximum mass found in all data: ", data.get_max_mass()
# time
time = data.get_time_list()
print "number of retention times: ", len(time)
print "retention time of 1st scan: ", time[0], "sec"
print "index of 400sec in time_list: ", data.get_index_at_time(400.0)
# TIC
tic = data.get_tic()
print "number of scans in TIC: ", len(tic)
print "start time of TIC: ", tic.get_time_at_index(0), "sec"
# raw scans
scans = data.get_scan_list()
print "number of masses in 1st scan: ", len(scans[0])