def remove_contaminants(data, parameters):
# type: (LFDataFrame, LFParameters) -> None
"""Remove straight m/z contaminants included in the contaminants CSV
file from input data.
Keyword Arguments:
data -- LFDataFrame instance
parameters -- LipidFinder's PeakFilter parameters instance
"""
mzCol = parameters['mzCol']
# Read the CSV file with the contaminants information
if (parameters['polarity'] == 'Negative'):
contaminants = pandas.read_csv(parameters['negContaminantsCSVPath'])
else:
contaminants = pandas.read_csv(parameters['posContaminantsCSVPath'])
# Remove every frame that matches with a known contaminant
toRemove = []
for index, mz in contaminants['MZ'].iteritems():
minMZ, maxMZ = mz_tol_range(mz, parameters['mzFixedError'],
parameters['mzPPMError'])
toRemove.extend(data[(minMZ <= data[mzCol])
& (data[mzCol] <= maxMZ)].index.tolist())
if (toRemove):
# Remove duplicates in the list to avoid errors
toRemove = list(set(toRemove))
data.drop('Contaminants removal', labels=toRemove, inplace=True)
# Reset the index of dataframe after the update
data.reset_index(inplace=True, drop=True)
def _collect_stack(array, index, rtGap, stackMZ, parameters):
# type: (numpy.ndarray, int, float, float, LFParameters) -> list
"""Get every feature that matches the stack m/z difference and
retention time gap from the previous feature to shape the
contaminant stack.
Keyword Arguments:
array -- array with every feature m/z, retention time (RT)
and index
index -- index of the previous feature in 'array'
rtGap -- RT difference between consecutive features
stackMZ -- contaminant m/z difference
parameters -- LipidFinder's PeakFilter parameters instance
"""
nextMZ = array[index, 0]
lastHitRT = array[index, 1]
rtDiff = 0
stackList = [index]
gapCount = 0
while (gapCount <= parameters['maxStackGap']):
# Calculate the expected m/z and RT of next stack feature
nextMZ += stackMZ
minMZ, maxMZ = mz_tol_range(nextMZ,
parameters['mzFixedError'],
parameters['mzPPMError'])
rtDiff += rtGap
expectedRT = lastHitRT + rtDiff
minRT, maxRT = rt_tol_range(expectedRT, parameters['maxRTDiffAdjFrame'])
matches = numpy.where(
(minMZ <= array[:, 0]) & (array[:, 0] <= maxMZ)
& (minRT <= array[:, 1]) & (array[:, 1] <= maxRT))[0]
if (len(matches) == 0):
gapCount += 1
else:
# Select the frame with the closest RT to the expected one
if (len(matches) == 1):
stackIndex = matches[0]
else:
stackIndex = matches[numpy.absolute(
array[matches, 1] - expectedRT).argmin()]
# Add the frame as member of the stack
stackList.append(stackIndex)
# Reset the information to calculate the next RT
lastHitRT = array[stackIndex, 1]
rtDiff = 0
# Reset the number of gaps
gapCount = 0
return (stackList)
def rm_full_frags(
array, # type: numpy.ndarray
fragments, # type: pandas.DataFrame
parameters # type: LFParameters
):
# type: (...) -> list[float]
"""Return an index list corresponding to common in-source fragments
in the given sample array.
Return the index list of all 'array' features that match the m/z
values provided in 'fragments' for which there is at least another
feature above the given m/z cut-off at the same retention time (RT).
All m/z and RT matching are computed within tolerance.
Keyword arguments:
array -- array with m/z, RT and index of the original
dataframe
fragments -- in-source fragments to be removed
parameters -- LipidFinder's PeakFilter parameters instance
"""
# Create an array with one in-source fragment m/z cut-off and m/z
# offset per row
fragsArray = numpy.stack(
(fragments['MZ'].values, fragments['MZCutOff'].values), axis=-1)
fragsIndex = []
for fragMZ, mzCutOff in fragsArray:
mzRange = mz_tol_range(fragMZ, parameters['mzFixedError'],
parameters['mzPPMError'])
# Get the index of 'array' features that match the in-source
# fragment m/z value
mzMatches = numpy.searchsorted(array[:, 0], mzRange)
if (mzMatches[0] == mzMatches[1]):
continue
for index in range(mzMatches[0], mzMatches[1]):
# To be a match, each feature must have the same RT
minRT, maxRT = rt_tol_range(array[index, 1], RT_TOLERANCE)
rtMatches = numpy.where((array[:, 0] >= mzCutOff)
& (array[:, 1] >= minRT)
& (array[:, 1] <= maxRT))[0]
if (len(rtMatches) > 0):
# Mark the feature as an in-source fragment
fragsIndex.append(index)
return fragsIndex
def amalgamate_data(negData, posData, parameters, dst=''):
# type: (object, object, LFParameters, str) -> None
"""Amalgamate negative and positive ion polarity dataframes.
'negData' and 'posData' have to match the same column layout as the
output files from LipidFinder's PeakFilter module.
For those frames with matching m/z and retention time, the one with
the lowest total intensity mean is discarded. Both files must have
the same column headings. If 'dst' is not an absolute path, the
current working directory will be used as starting point. If
"amalgamated.csv" file already exists, it will be overwritten.
Keyword Arguments:
negData -- negative polarity LFDataFrame or pandas.DataFrame
instance
posData -- positive polarity LFDataFrame or pandas.DataFrame
instance
parameters -- LipidFinder's Amalgamator parameters instance
dst -- destination directory where the log file and the
amalgamated data CSV file will be saved
[default: current working directory]
"""
# Set the log file where the information about the steps performed
# is saved
logFilePath = 'amalgamator.log'
if (dst):
logFilePath = os.path.join(dst, logFilePath)
# Create logger and its file handler
logger = logging.getLogger()
logger.setLevel(logging.INFO)
handler = logging.FileHandler(logFilePath)
handler.setLevel(logging.INFO)
formatter = logging.Formatter('[%(asctime)s] %(message)s')
handler.setFormatter(formatter)
logger.addHandler(handler)
# Write initial information in log file
logger.info(("Starting Amalgamator. Negative dataframe has %d rows and "
"Positive dataframe has %d rows."), len(negData.index),
len(posData.index))
mzCol = parameters['mzCol']
rtCol = parameters['rtCol']
# Check if columns in both dataframes are the same
if (set(negData.columns) != set(posData.columns)):
diffCols = set(negData.columns).symmetric_difference(posData.columns)
raise IOError(("Input dataframes do not share the same column names: "
"{0}").format(', '.join(diffCols)))
# Check for misspelling errors in m/z or retention time column names
if ((mzCol not in negData.columns) or (rtCol not in negData.columns)):
raise KeyError("Missing '{0}' or '{1}' column(s)".format(mzCol, rtCol))
# Get the indices for intensity columns
firstIndex = parameters['firstSampleIndex'] - 1
lastIndex = firstIndex + parameters['numSamples']
# Calculate the mean of every non-zero value of the mean columns of
# each input dataframe and round it to the nearest integer. Replace
# any NaN output from mean() by zero.
totalMean = lambda x: numpy.rint(
numpy.nan_to_num(x[numpy.where(x > 0)[0]].mean())).astype(int)
negData['TotalMean'] = negData.iloc[:,
firstIndex:lastIndex].apply(totalMean,
axis=1)
posData['TotalMean'] = posData.iloc[:,
firstIndex:lastIndex].apply(totalMean,
axis=1)
nind = negData.index.values
nmz = negData[mzCol].values
nrt = negData[rtCol].values
nmeans = negData['TotalMean'].values
negCol = list(negData.columns.values)
posCol = list(posData.columns.values)
# Empty results dataframe
results = pandas.DataFrame(columns=negCol)
polColIndex = results.columns.get_loc('Polarity')
# Start progress bar
progress = 0
total = len(nind) + 1
print_progress_bar(progress, total, prefix='Amalgamator progress:')
# Loop through indices in negative file
for i in nind:
# Update progress bar
progress += 1
print_progress_bar(progress, total, prefix='Amalgamator progress:')
negMass = nmz[i]
negRT = nrt[i]
pmz = posData[mzCol].values
prt = posData[rtCol].values
pmeans = posData['TotalMean'].values
negMassH2 = negMass + HYDROGEN
mzRange = mz_tol_range(negMassH2, parameters['mzFixedError'],
parameters['mzPPMError'])
rtRange = rt_tol_range(negRT, parameters['maxRTDiffAdjFrame'])
matchesH2 = list(
numpy.where((pmz >= mzRange[0]) & (pmz <= mzRange[1])
& (prt >= rtRange[0])
& (prt <= rtRange[1]))[0])
# First, look for H2 matches
if (matchesH2):
indMatch = __bestMatch__(matchesH2, negMassH2, pmz, negRT, prt,
parameters)
# Keep the frame with the highest total mean
if (pmeans[indMatch] > nmeans[i]):
results = results.append(posData.iloc[indMatch],
ignore_index=True)
if (parameters['combineIntensities']):
results.iloc[-1, firstIndex : lastIndex] = \
results.iloc[-1, firstIndex : lastIndex] \
+ negData.iloc[i, firstIndex : lastIndex]
results.iloc[-1, polColIndex] += ' (Combined)'
else:
results.iloc[-1, polColIndex] += ' (Both)'
else:
results = results.append(negData.iloc[i], ignore_index=True)
if (parameters['combineIntensities']):
results.iloc[-1, firstIndex : lastIndex] = \
results.iloc[-1, firstIndex : lastIndex] \
+ posData.iloc[indMatch, firstIndex : lastIndex]
results.iloc[-1, polColIndex] += ' (Combined)'
else:
results.iloc[-1, polColIndex] += ' (Both)'
logger.info('Match found: Negative ID %d - Positive ID %d.',
negData.iloc[i, 0], posData.iloc[indMatch, 0])
# Remove match from positive dataframe, avoiding writing
# the action to the log file
if (isinstance(posData, LFDataFrame)):
super(LFDataFrame, posData).drop(indMatch, inplace=True)
else:
posData.drop(indMatch, inplace=True)
posData.reset_index(inplace=True, drop=True)
pmz = posData[mzCol].values
prt = posData[rtCol].values
pmeans = posData['TotalMean'].values
continue
# If there are no H2 matches, look for CH4 matches
negMassCH4 = negMass + METHANE
mzRange = mz_tol_range(negMassCH4, parameters['mzFixedError'],
parameters['mzPPMError'])
matchesHCH3 = list(
numpy.where((pmz >= mzRange[0]) & (pmz <= mzRange[1])
& (prt >= rtRange[0])
& (prt <= rtRange[1]))[0])
if (matchesHCH3):
indMatch = __bestMatch__(matchesHCH3, negMassCH4, pmz, negRT, prt,
parameters)
# Keep the frame with the highest total mean
if (pmeans[indMatch] > nmeans[i]):
results = results.append(posData.iloc[indMatch],
ignore_index=True)
if (parameters['combineIntensities']):
results.iloc[-1, firstIndex : lastIndex] = \
results.iloc[-1, firstIndex : lastIndex] \
+ negData.iloc[i, firstIndex : lastIndex]
results.iloc[-1, polColIndex] += ' (Combined)'
else:
results.iloc[-1, polColIndex] += ' (Both)'
else:
results = results.append(negData.iloc[i], ignore_index=True)
if (parameters['combineIntensities']):
results.iloc[-1, firstIndex : lastIndex] = \
results.iloc[-1, firstIndex : lastIndex] \
+ posData.iloc[indMatch, firstIndex : lastIndex]
results.iloc[-1, polColIndex] += ' (Combined)'
else:
results.iloc[-1, polColIndex] += ' (Both)'
logger.info('Match found: Negative ID %d - Positive ID %d.',
negData.iloc[i, 0], posData.iloc[indMatch, 0])
# Remove match from positive dataframe, avoiding writing
# the action to the log file
if (isinstance(posData, LFDataFrame)):
super(LFDataFrame, posData).drop(indMatch, inplace=True)
else:
posData.drop(indMatch, inplace=True)
posData.reset_index(inplace=True, drop=True)
pmz = posData[mzCol].values
prt = posData[rtCol].values
pmeans = posData['TotalMean'].values
continue
results = results.append(negData.iloc[i], ignore_index=True)
# Append what remains in the positive dataframe (unmatched positive
# m/z values)
results = results.append(posData, ignore_index=True)
if (pandas.__version__ < '0.23.0'):
# Fix unexpected column sorting from append() in pandas v0.20.3
# or newer (solved in v0.23.0 with argument "sort=False")
results = results.reindex(negCol, axis=1)
results.drop('TotalMean', axis=1, inplace=True)
# Sort results by m/z and retention time and create the CSV file
results.sort_values([mzCol, rtCol], inplace=True, kind='mergesort')
results.to_csv(os.path.join(dst, 'amalgamated.csv'), index=False)
# Update progress bar
print_progress_bar(total, total, prefix='Amalgamator progress:')
# Write the final information in log file and remove handler
logger.info('Amalgamator completed. Output dataframe has %d rows.\n',
len(results.index))
handler.close()
logger.removeHandler(handler)
def _detect_sample_isotopes(array, parameters):
"""Return an array with the tagged parents and their corresponding
isotopes in the same order as in the given sample array.
Keyword Arguments:
array -- array with m/z, retention time (RT), sample's
intensity mean and index of the original dataframe
parameters -- LipidFinder's PeakFilter parameters instance
"""
# Get the corresponding symbol for the polarity of the data (+ or -)
polSign = '+' if (parameters['polarity'].lower() == 'positive') else '-'
# Create an array of empty strings that will contain the tagged
# parents and their corresponding isotopes
tagArray = numpy.full(len(array), '', dtype=object)
# Loop over each m/z to search for isotopes
isotopesIndex = set()
for index in range(0, len(array)):
# Skip if frame has already been identified as an isotope
if (array[index, 3] in isotopesIndex):
continue
for isoPeak in range(1, parameters['numIsotopes'] + 1):
parentMZ = array[index, 0]
tagID = int(array[index, 3])
# Get the first and last indexes of the frames that are
# within the first isotope m/z range for the current analyte
isotopeMZ = parentMZ + ISO_OFFSET * isoPeak
minMZ, maxMZ = mz_tol_range(isotopeMZ, parameters['mzFixedError'],
parameters['mzPPMError'])
mzMatches = numpy.searchsorted(array[:, 0], [minMZ, maxMZ])
if (mzMatches[0] == mzMatches[1]):
# Have not found any analyte with an isotope-like m/z
if (isoPeak == 1):
# The first isotope must exists to search for others
break
else:
continue
# Filter m/z matches with the same RT as the parent
parentRT = array[index, 1]
minRT, maxRT = rt_tol_range(parentRT,
parameters['maxRTDiffAdjFrame'])
rtMatches = numpy.where(
(array[mzMatches[0] : mzMatches[1], 1] >= minRT)
& (array[mzMatches[0] : mzMatches[1], 1] <= maxRT))[0]
if (len(rtMatches) == 0):
# No candidates are within the same RT
if (isoPeak == 1):
# The first isotope must exists to search for others
break
else:
continue
# Resultant indexes are based on the previous search
rtMatches += mzMatches[0]
# Filter the candidate isotopes by intensity
parentInten = array[index, 2]
# The intensity range coefficients vary depending on the
# isotope number
if (isoPeak == 1):
# Get an estimated maximum number of C in the molecule
numC = round(parentMZ / 12)
# Calculate isotopic distribution based on polynomial
# expansion
baseIntensity = parentInten * (numC ** 1.3) * 0.002
minIntensity = baseIntensity * parameters['isoIntensityCoef'][0]
maxIntensity = baseIntensity * parameters['isoIntensityCoef'][1]
elif (isoPeak == 2):
# Get an estimated maximum number of C in the molecule
numC = round(parentMZ / 12)
# Calculate isotopic distribution based on polynomial
# expansion
baseIntensity = parentInten * (numC ** 1.7) * 0.0001
minIntensity = baseIntensity * parameters['isoIntensityCoef'][0]
maxIntensity = baseIntensity * parameters['isoIntensityCoef'][1]
else:
# Calculate isotopic distribution with the same formula
# as CAMERA (from XCMS)
minIntensity = parentInten * float('1e-{0}'.format(isoPeak + 2))
maxIntensity = parentInten * 2
isotopes = numpy.where((array[rtMatches, 2] >= minIntensity)
& (array[rtMatches, 2] <= maxIntensity))[0]
if (len(isotopes) == 0):
# No candidates have an intensity within expected range
if (isoPeak == 1):
# The first isotope must exists to search for others
break
else:
continue
# Resultant indexes are based on the previous search
isotopes += rtMatches[0]
# Tag the analyte as isotope and save its index to avoid
# checking it as parent of other analytes
tagArray[isotopes] = '[{0}][M+{1}]{2}'.format(tagID, isoPeak,
polSign)
isotopesIndex.update(array[isotopes, 3])
else:
# Tag the analyte as parent
tagArray[index] = '[{0}][M]{1}'.format(tagID, polSign)
return tagArray
def remove_stacks(data, parameters):
# type: (LFDataFrame, LFParameters) -> None
"""Detect lipid and contaminant stacks and delete all ions present
(in lipid stacks the parent is retained).
A stack is a series of ions differing in m/z by a user-defined fixed
mass shift. Lipid stacks elute at same retention time (RT) whilst
contaminant stacks increase their RT as overall m/z increases.
Firstly, the m/z is checked for a lipid stack and if a stack is
present, all ions except the parent are deleted and the next m/z is
checked. If no lipid stack is found then the m/z is checked for
contaminant stacks. If found, the whole stack including the parent
is removed. The list of lipid and contaminant stack mass differences
is imported from the stacks CSV file.
Keyword Arguments:
data -- LFDataFrame instance
parameters -- LipidFinder's PeakFilter parameters instance
"""
mzCol = parameters['mzCol']
rtCol = parameters['rtCol']
firstSample = parameters['firstSampleIndex'] - 1
lastSample = firstSample \
+ (parameters['numSamples'] * parameters['numTechReps'])
# Read the CSV file with the stacks information
stacks = pandas.read_csv(parameters['stacksCSVPath'])
# Separate lipid and contaminant stacks
lipidStacksMZ = stacks.loc[stacks['Category'] == 'Lipid', 'MZ'].values
contStacksMZ = stacks.loc[stacks['Category'] == 'Contaminant', 'MZ'].values
# Build an array with m/z, RT and index to track the frames removed
# as part of a stack
array = numpy.stack((data[mzCol].values, data[rtCol].values,
data.index.values), axis=-1)
# Start the loop to find every stack in the dataset
parentIndex = 0
toRemove = {'lipid': [], 'contam': []}
while (parentIndex < (len(array) - 1)):
parentMZ, parentRT = array[parentIndex, 0:2]
# Lipid stack removal where m/z and RT have to be an exact match
# (within tolerance)
minRT, maxRT = rt_tol_range(parentRT, parameters['maxRTDiffAdjFrame'])
# Stacks can be of only one type: if a lipid stack is found the
# parent m/z will not be analysed as part of a contaminant stack
for stackMZ in lipidStacksMZ:
stackDiff = 0
gapCount = 0
stackList = []
while (gapCount <= parameters['maxStackGap']):
# Calculate the expected m/z of the next feature
stackDiff += stackMZ
minMZ, maxMZ = mz_tol_range(parentMZ + stackDiff,
parameters['mzFixedError'],
parameters['mzPPMError'])
matches = numpy.where(
(minMZ <= array[:, 0]) & (array[:, 0] <= maxMZ)
& (minRT <= array[:, 1]) & (array[:, 1] <= maxRT))[0]
if (len(matches) == 0):
gapCount += 1
else:
# Select the feature with the closest RT to parent
if (len(matches) == 1):
stackIndex = matches[0]
else:
stackIndex = matches[numpy.absolute(
array[matches, 1] - parentRT).argmin()]
# Add the frame as member of the stack
stackList.append(stackIndex)
if (len(stackList) >= MIN_LIPID_STACK):
# Mark frames to be removed as part of the lipid stack
indexes = array[stackList, 2].tolist()
toRemove['lipid'].extend(indexes)
# Remove frames in stack from array
array = numpy.delete(array, stackList, axis=0)
if (parameters['lipidStackAddition']):
# Add stack intensities to parent frame
data.iloc[parentIndex, firstSample : lastSample] += \
data.iloc[indexes, firstSample : lastSample].sum(
axis=0)
break
else:
# No lipid stacks where found: search for contamination
# stacks where m/z has to be an exact match (within
# tolerance) and the RT between every two consecutive
# features is the same (in accordance with the m/z
# difference)
for stackMZ in contStacksMZ:
stackDiff = 0
stackList = []
# The RT gap is unknown until the stack has two
# features, so calculate the minimum RT of next feature
minRT = rt_tol_range(parentRT,
parameters['maxRTDiffAdjFrame'])[1]
# Get every possible stack within the maximum gap
# distance and keep the largest one
for gapCount in range(0, parameters['maxStackGap']):
# Calculate the expected m/z of next stack feature
stackDiff += stackMZ
minMZ, maxMZ = mz_tol_range(parentMZ + stackDiff,
parameters['mzFixedError'],
parameters['mzPPMError'])
matches = numpy.where(
(minMZ <= array[:, 0]) & (array[:, 0] <= maxMZ)
& (minRT < array[:, 1]))[0]
# Explore every possible stack (different RT gap)
# and keep the largest one
for i in range(0, len(matches)):
nextMZ, nextRT = array[matches[i], 0:2]
rtGap = (nextRT - parentRT) / (gapCount + 1)
matchStackList = _collect_stack(
array, matches[i], rtGap, stackMZ, parameters)
if (len(matchStackList) > len(stackList)):
stackList = matchStackList
if ((len(stackList) + 1) >= MIN_CONTAM_STACK):
# Mark frames to be removed as part of the
# contaminant stack (parent included)
stackList.append(parentIndex)
indexes = array[stackList, 2].tolist()
toRemove['contam'].extend(indexes)
# Remove frames in stack from array
array = numpy.delete(array, stackList, axis=0)
# Adjust index after parent has been removed
parentIndex -= 1
break
parentIndex += 1
# Remove lipid and/or contaminant stack features
if (toRemove['lipid'] or toRemove['contam']):
data.drop('Stacks removal (lipid)', labels=toRemove['lipid'],
inplace=True)
data.drop('Stacks removal (contaminant)', labels=toRemove['contam'],
inplace=True)
# Reset the index of dataframe after the update
data.reset_index(inplace=True, drop=True)
def __rep_adduct_removal__(replicate, # pandas.Series
adductsPairs, # list
adducts, # pandas.DataFrame
mzArray, # numpy.array
rtArray, # numpy.array
parameters # LFParameters
):
# type: (...) -> pandas.Series
"""Detect pairs of adducts in the given sample replicate and set to
zero the lowest intensity of each pair.
The m/z and retention time (RT) matches are done within a tolerance.
Keyword Arguments:
replicate -- replicate's intensities
pairs -- list of paired adducts
adducts -- adducts information
mzArray -- sample replicate's m/z values
rtArray -- sample replicate's rt values
parameters -- LipidFinder's PeakFilter parameters instance
"""
# Get the index of all intensities that are not zero
nonZeroIndices = replicate.values.nonzero()[0]
# Create an array of intensities that are not zero, and the arrays
# with their corresponding m/z and RT values
nzIntensities = numpy.copy(replicate.values[nonZeroIndices])
nzMZ = numpy.copy(mzArray[nonZeroIndices])
nzRT = numpy.copy(rtArray[nonZeroIndices])
# Create an array to hold a reference to the first adduct found
# (default: "")
adductTags = numpy.empty_like(nonZeroIndices, dtype=str)
adductTags.fill('')
for pair in adductsPairs:
# Get the adducts information of the pair to create the lambda
# function to calculate the offset of the given mass
pairInfo = adducts.loc[adducts.iloc[:, 0].isin(pair)]
# abs() function makes the pairs order insensitive
get_offset = lambda x: \
abs(x - (pairInfo.iloc[1, 1] * (x - pairInfo.iloc[0, 2]) /
pairInfo.iloc[0, 1] + pairInfo.iloc[1, 2]))
index = 0
while (index < (nonZeroIndices.size - 1)):
# Get first adduct tag of the source index
tag = adductTags[index]
# If source frame has been previously tagged and tag is not
# equal to lower mass species then this frame cannot be
# considered further so consider next frame
if (tag and (tag != pair[0])):
index += 1
continue
# Get the m/z and RT of the current frame
mz = nzMZ[index]
rt = nzRT[index]
# The adduct mass for the source mass with current pair
adductMZ = mz + get_offset(mz)
# The limits of a mass that could be an adduct of the source
# mass, including the error tolerance
minAdductMZ, maxAdductMZ = mz_tol_range(
adductMZ, parameters['mzFixedError'],
parameters['mzPPMError'])
# Get the tolerance range for the RT
minRT, maxRT = rt_tol_range(rt, parameters['maxRTDiffAdjFrame'])
# Create a dataframe of potential adducts by m/z and RT
potentialAdducts = numpy.where(
(nzMZ >= minAdductMZ) & (nzMZ <= maxAdductMZ)
& (nzRT >= minRT) & (nzRT <= maxRT))[0]
if (potentialAdducts.size > 0):
# Get the index of the adduct with the closest RT to the
# subject RT
adductIndex = potentialAdducts[numpy.absolute(
nzRT[potentialAdducts] - rt).argmin()]
# If the intensity of the adduct is greater than the
# intensity of the source, set the latter to 0.
# Otherwise, set the former to 0.
if (nzIntensities[adductIndex] > nzIntensities[index]):
# Go ahead if the adduct has not been tagged yet
if (not adductTags[adductIndex]):
# Record adduct species
adductTags[adductIndex] = pairInfo.iloc[1, 0]
if (parameters['adductAddition']):
replicate.values[nonZeroIndices[adductIndex]] += \
nzIntensities[index]
# Set source intensity to 0
replicate.values[nonZeroIndices[index]] = 0
# Remove source from each array
nonZeroIndices = numpy.delete(nonZeroIndices, index)
nzIntensities = numpy.delete(nzIntensities, index)
nzMZ = numpy.delete(nzMZ, index)
nzRT = numpy.delete(nzRT, index)
adductTags = numpy.delete(adductTags, index)
# 'index' will now point to the next frame
continue
else:
if (not tag):
# Record adduct species
adductTags[index] = pairInfo.iloc[0, 0]
if (parameters['adductAddition']):
replicate.values[nonZeroIndices[index]] += \
nzIntensities[adductIndex]
# Set adduct intensity to 0
replicate.values[nonZeroIndices[adductIndex]] = 0
# Remove adduct from each array
nonZeroIndices = numpy.delete(nonZeroIndices, adductIndex)
nzIntensities = numpy.delete(nzIntensities, adductIndex)
nzMZ = numpy.delete(nzMZ, adductIndex)
nzRT = numpy.delete(nzRT, adductIndex)
adductTags = numpy.delete(adductTags, adductIndex)
index += 1
return replicate
def rm_neutral_loss_frags(
array, # type: numpy.ndarray
losses, # type: pandas.DataFrame
parameters # type: LFParameters
):
# type: (...) -> list[float]
"""Return an index list corresponding to the features in the given
sample array that have been fragmented.
Return the index list of all 'array' features that have lost one of
the m/z in 'losses' and their complete counterpart is present in the
data. The features to be removed must be higher than the given
cut-off. All m/z and retention time (RT) matching are computed
within tolerance.
Keyword arguments:
array -- array with m/z, RT and index of the original
dataframe
losses -- neutral losses to subtract in order to detect
fragmented features
parameters -- LipidFinder's PeakFilter parameters instance
"""
# Create an array with one m/z cut-off and neutral loss m/z per row
fragsArray = numpy.stack((losses['MZCutOff'].values, losses['MZ'].values),
axis=-1)
# Create a dictionary with cut-off values as keys and their
# corresponding neutral loss m/z in lists as values
fragsDict = {}
for mzCutOff, mzLoss in fragsArray:
fragsDict.setdefault(mzCutOff, []).append(mzLoss)
matchIndexSet = set()
for mzCutOff in viewkeys(fragsDict):
# Get the index of the first m/z value in 'array' greater than
# the m/z cut-off
firstIndex = numpy.searchsorted(array[:, 0], mzCutOff)
for index in range(firstIndex, len(array)):
for mzLoss in fragsDict[mzCutOff]:
# Look for in-source fragments, that is, features that
# are the result of subtracting the neutral loss to the
# parent's m/z and elute at the same RT
fragMZ = array[index, 0] - mzLoss
mzRange = mz_tol_range(fragMZ, parameters['mzFixedError'],
parameters['mzPPMError'])
# Get first and last indexes of the features within the
# m/z range
mzMatches = numpy.searchsorted(array[:, 0], mzRange)
if (mzMatches[0] == mzMatches[1]):
continue
# In order to be considered a match, each feature must
# have the same RT
minRT, maxRT = rt_tol_range(array[index, 1], RT_TOLERANCE)
rtMatches = numpy.where(
(array[mzMatches[0]:mzMatches[1], 1] >= minRT)
& (array[mzMatches[0]:mzMatches[1], 1] <= maxRT))[0]
if (len(rtMatches) == 0):
continue
# The resultant indexes are based on the starting index
# of the search ('mzMatches[0]')
rtMatches += mzMatches[0]
# The union of sets will handle any index repetition
matchIndexSet.update(set(rtMatches))
return list(matchIndexSet)
