def rawMassLynx_MS_load(path=None, inputFile='output.1dMZ', normalize=True, **kwargs):
datapath = os.path.join(path, inputFile)
# Load data into array
msData = np.fromfile(datapath, dtype=np.float32)
msDataClean = msData[2:(-1)]
numberOfRows = int(len(msDataClean) / 2)
msDataReshape = msDataClean.reshape((2, numberOfRows), order='F')
# Extract MS data
msX = msDataReshape[1]
msY = msDataReshape[0]
# clean-up MS file
firstBad = strictly_increasing(msX)
if firstBad == True: pass
else:
firstBadIdx = np.where(msX == firstBad)
try:
msX = msX[0:(firstBadIdx[0][0] - 1)]
msY = msY[0:(firstBadIdx[0][0] - 1)]
except IndexError: pass
# clean-up filepath
try: clean_up(datapath)
except: pass
# Normalize MS data
if normalize:
msY = msY / max(msY)
return msX, msY
def rawMassLynx_2DT_load(path=None, inputFile='output.2dRTDT', normalize=False, **kwargs):
datapath = os.path.join(path, inputFile)
imsData = np.fromfile(datapath, dtype=np.int32)
imsDataClean = imsData[3::]
numberOfRows = imsData[1]
imsDataReshape = imsDataClean.reshape((200, numberOfRows), order='F') # Reshapes the list to 2D array
imsDataSplit = np.hsplit(imsDataReshape, int(numberOfRows / 5)) # Splits the array into [scans,200,5] array
imsDataSplit = np.sum(imsDataSplit, axis=2).T # Sums each 5 RT scans together
"""
There is a bug - sometimes when extracting, the last column values
go way outside of range and make the plot look terrible. Hacky way
round this is to check that any values fall outside the range and if so,
set the last column to 0s...
"""
# Test to ensure all values are above 0 or below 1E8
for value in imsDataSplit[:, -1]:
if value < 0:
imsDataSplit[:, -1] = 0
elif value > 10000000:
imsDataSplit[:, -1] = 0
# clean-up filepath
try: clean_up(datapath)
except: pass
if normalize:
imsDataSplitNorm = normalize(imsDataSplit.astype(np.float64), axis=0, norm='max') # Norm to 1
return imsDataSplit, imsDataSplitNorm
else:
return imsDataSplit
def rawMassLynx_MZDT_load(path=None, inputFile='output.2dDTMZ', normalize=False, **kwargs):
datapath = os.path.join(path, inputFile)
imsData = np.fromfile(datapath, dtype=np.int32)
imsDataClean = imsData[3::]
numberOfBins = imsData[0]
imsDataSplit = imsDataClean.reshape((200, numberOfBins), order='C') # Reshapes the list to 2D array
# clean-up filepath
try: clean_up(datapath)
except: pass
if normalize:
imsDataSplitNorm = normalize(imsDataSplit.astype(np.float64), axis=0, norm='max') # Norm to 1
return imsDataSplit, imsDataSplitNorm
else:
return imsDataSplit
def rawMassLynx_RT_load(path=None, inputFile='output.1dRT', normalize=False, **kwargs):
datapath = os.path.join(path, inputFile)
rtData = np.fromfile(datapath, dtype=np.int32)
rtDataClean = rtData[3::]
nScans = rtData[1]
rtDataReshape = rtDataClean.reshape((nScans, 2), order='C')
rtData1D = rtDataReshape[:, 1]
# clean-up filepath
try: clean_up(datapath)
except: pass
if normalize:
rtData1DNorm = rtData1D.astype(np.float64) / max(rtData1D)
return rtData1D, rtData1DNorm
else:
return rtData1D
def rawMassLynx_DT_load(path=None, inputFile='output.1dDT', normalize=False, **kwargs):
"""
Load data for 1D IM-MS data
"""
datapath = os.path.join(path, inputFile)
imsData = np.fromfile(datapath, dtype=np.int32)
imsDataClean = imsData[3::]
imsDataReshape = imsDataClean.reshape((200, 2), order='C')
imsData1D = imsDataReshape[:, 1]
# clean-up filepath
try: clean_up(datapath)
except: pass
if normalize:
imsData1DNorm = imsData1D.astype(np.float64) / max(imsData1D)
return imsData1DNorm
else:
return imsData1D