def detect_one_run(run, args):
infile = os.path.join(args.indir, run)
print "processing GC-MS file:", infile
# sys.stdout("processing GCSM run:", run)
# load the input GC-MS file
try:
if args.ftype == 'CDF':
from pyms.GCMS.IO.ANDI.Function import ANDI_reader
data = ANDI_reader(infile)
elif args.ftype == 'JDX':
#data = JCAMP_reader(in_file)
data = pyms.GCMS.IO.JCAMP.Function.JCAMP_OpenChrom_reader(infile)
else:
raise ValueError('can only load ANDI (CDF) or JDX files!')
except:
print "Failure to load input file ", infile
else:
data.trim(args.trimstart + "m", args.trimend + "m")
# get TIC. Would prefer to get from smoothed IM but API is faulty!
tic = data.get_tic()
# integer mass
im = build_intensity_matrix_i(data)
# would be nice to do noise_mult*noise_level using the noise level AFTER smoothing,
# but i can't seem to get the TIC for the smoothed IM.
peak_list = call_peaks(im, tic, True, args)
return peak_list, run
def detect_one_run(run, args):
infile = os.path.join(args.indir, run)
print "processing GC-MS file:", infile
# sys.stdout("processing GCSM run:", run)
# load the input GC-MS file
try:
if args.ftype == 'CDF':
from pyms.GCMS.IO.ANDI.Function import ANDI_reader
data = ANDI_reader(infile)
elif args.ftype == 'JDX':
#data = JCAMP_reader(in_file)
data = pyms.GCMS.IO.JCAMP.Function.JCAMP_OpenChrom_reader(infile)
else:
raise ValueError('can only load ANDI (CDF) or JDX files!')
except:
print "Failure to load input file ", infile
else:
data.trim(args.trimstart+"m",args.trimend+"m")
# get TIC. Would prefer to get from smoothed IM but API is faulty!
tic = data.get_tic()
# integer mass
im = build_intensity_matrix_i(data)
# would be nice to do noise_mult*noise_level using the noise level AFTER smoothing,
# but i can't seem to get the TIC for the smoothed IM.
peak_list = call_peaks(im, tic, True, args)
return peak_list, run
def load_run(infile):
try:
if args.ftype == 'CDF':
from pyms.GCMS.IO.ANDI.Function import ANDI_reader
data = ANDI_reader(infile)
elif args.ftype == 'JDX':
#data = JCAMP_reader(in_file)
data = pyms.GCMS.IO.JCAMP.Function.JCAMP_OpenChrom_reader(infile)
else:
raise ValueError('can only load ANDI (CDF) or JDX files!')
except:
print "Failure to load input file ", infile
else:
data.trim("4.0m", "20.0m")
# get TIC. Would prefer to get from smoothed IM but API is faulty!
tic = data.get_tic()
# integer mass
return build_intensity_matrix_i(data), tic
def load_run(infile):
try:
if args.ftype == 'CDF':
from pyms.GCMS.IO.ANDI.Function import ANDI_reader
data = ANDI_reader(infile)
elif args.ftype == 'JDX':
#data = JCAMP_reader(in_file)
data = pyms.GCMS.IO.JCAMP.Function.JCAMP_OpenChrom_reader(infile)
else:
raise ValueError('can only load ANDI (CDF) or JDX files!')
except:
print "Failure to load input file ", infile
else:
data.trim("4.0m", "20.0m")
# get TIC. Would prefer to get from smoothed IM but API is faulty!
tic = data.get_tic()
# integer mass
return build_intensity_matrix_i(data), tic
def missing_peak_finder(sample, andi_file, points=7, null_ions=[73, 207],\
crop_ions=[45,300], threshold=100000, rt_window=10):
"""
@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(), "andi_file:", andi_file
data = ANDI_reader(andi_file)
# 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():
#JT: Debug peak attributes
#attrs = vars(mp)
#print ', '.join("%s: %s" % item for item in attrs.items())
mp_rt = mp.get_rt()
#print(repr(mp_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')
from pyms.Deconvolution.BillerBiemann.Function import BillerBiemann, \
rel_threshold, num_ions_threshold
from pyms.Experiment.Class import Experiment
from pyms.Experiment.IO import store_expr
# deconvolution and peak list filtering parameters
points = 9; scans = 2; n = 3; t = 3000; r = 2;
andi_file = "/x/PyMS/data/a0806_077.cdf"
data = ANDI_reader(andi_file)
# integer mass
im = build_intensity_matrix_i(data)
# 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)
ic_smooth = savitzky_golay(ic1)
ic_base = tophat(ic_smooth, struct="1.5m")
im.set_ic_at_index(ii, ic_base)
# do peak detection on pre-trimmed data
# get the list of Peak objects
from pyms.Baseline.TopHat import tophat
from pyms.Display.Class import Display
from pyms.Peak.Function import peak_sum_area
from pyms.Peak.IO import store_peaks
from pyms.Deconvolution.BillerBiemann.Function import BillerBiemann, \
rel_threshold, num_ions_threshold
from pyms.Simulator.Function import gcms_sim, add_gaussc_noise
# read in raw data
andi_file = "/x/PyMS/data/gc01_0812_066.cdf"
data = ANDI_reader(andi_file)
data.trim(4101, 4350)
# Build Intensity Matrix
real_im = build_intensity_matrix_i(data)
n_scan, n_mz = real_im.get_size()
# perform necessary pre filtering
for ii in range(n_mz):
ic = real_im.get_ic_at_index(ii)
ic_smooth = savitzky_golay(ic)
ic_bc = tophat(ic_smooth, struct="1.5m")
real_im.set_ic_at_index(ii, ic_bc)
# Detect Peaks
peak_list = BillerBiemann(real_im, points=3, scans=2)
print "Number of peaks found in real data: ", len(peak_list)
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')
