def testStudyVariance(self):
alnFileName = "../../alignmentFiles/mmult.list-FINAL.aln"
workingDirectory = "../../alignmentFiles/"
resultsFileName2 = "../../alignmentFiles/results2.results"
generateMultipleAlignment.getResidues(0.8, resultsFileName2, alnFileName, workingDirectory, False, True, False)
AtomicresultsFileName = "../../alignmentFiles/results2.Atomic"
generateMultipleAlignment.composeAtomicMatrix(resultsFileName2, AtomicresultsFileName)
filteredAtomicMatrixFileName = "../../alignmentFiles/filteredResults2.Atomic"
generateMultipleAlignment.filterAtomicMatrix(AtomicresultsFileName, filteredAtomicMatrixFileName)
caMatrixFileName = "../../alignmentFiles/filteredResults2.Atomic"
dimensions = generateMultipleAlignment.getDimensions(caMatrixFileName)
(mask, matrix) = svdAnalysis.loadCAmatrix(caMatrixFileName, dimensions)
(caMatrix, caMask) = svdAnalysis.getCovarianceMatrix(mask, matrix)
(atoms_variance, atoms_variance_mask) = svdAnalysis.getAtomsVariance(caMatrix, caMask)
(max, min, mean, normalizedAtoms) = svdAnalysis.atoms_variance_statistics(atoms_variance, atoms_variance_mask)
print ("max %f \n" % (max,))
print ("min %f \n" % (min,))
print ("mean %f \n" % (mean,))
print normalizedAtoms
print "\n\n"
workingDirectory = "../../alignmentFiles/"
varianceFileName = workingDirectory + "variance.atomic"
maskFileName = workingDirectory + "variance_mask.atomic"
svdAnalysis.serializeAtomsVariance(atoms_variance, atoms_variance_mask, varianceFileName, maskFileName)
(atoms_variance, atoms_variance_mask) = svdAnalysis.encarnateAtomsVariance(varianceFileName, maskFileName)
assert 30 == len(atoms_variance[0])
for variance in atoms_variance[0]:
print ("%3.f \n" % (variance,))
def __load_variability_data(self, minimumPercentage):
"""
__alphaCarbonsFileName where to write the aligned C_alphas for each pdb.
__alnFileName the alignment filename. No need to be already formatted
__workingDirectory where the pdbs are stored
onlyBackbone select only the backbone. Default False
onlyAlphaCarbon select only the alphaCarbons. Default True. Consider that all the latter programs work over this option and no other!!!!
self.removedLines contains the removed lines, that is if a residue position is not aligned over a percentage for all
the pdbs, this line is removed, and its order is this array
"""
self.__mmult = generateMultipleAlignment.getResidues(minimumPercentage, self.__alphaCarbonsFileName, self.__alnFileName,
self.__workingDirectory, False, True, False)
generateMultipleAlignment.composeAtomicMatrix(self.__alphaCarbonsFileName, self.__rawAtomicMatrixFileName)
self.removedLines = generateMultipleAlignment.filterAtomicMatrix(self.__rawAtomicMatrixFileName, self.__filteredAtomicMatrixFileName)
self.dimensions = generateMultipleAlignment.getDimensions(self.__filteredAtomicMatrixFileName)
def testGlobal(self):
caMatrixFileName = "../../alignmentFiles/filteredResults2.Atomic"
dimensions = generateMultipleAlignment.getDimensions(caMatrixFileName)
print ("%d %d\n" % (dimensions[0], dimensions[1]))
(mask, matrix) = svdAnalysis.loadCAmatrix(caMatrixFileName, dimensions)
(caMatrix, caMask) = svdAnalysis.getCovarianceMatrix(mask, matrix)
print ("%d %d\n" % (caMatrix.shape[0], caMatrix.shape[1]))
matrixFileName = "../../alignmentFiles/matrix.float"
maskFileName = "../../alignmentFiles/mask.float"
# print("\n caMatrix\n")
# print caMatrix
# print("\ncaMask\n")
# print caMask
svdAnalysis.serialize_matrix(caMatrix, matrixFileName)
diagonalFileName = "../../alignmentFiles/diagonal.float"
(ca_atoms, ca_mask) = svdAnalysis.getAtomsVariance(caMatrix, caMask)
svdAnalysis.serialize_matrix(ca_atoms, diagonalFileName)
svdAnalysis.serialize_matrix(ca_mask, maskFileName)
def testGlobalReal(self):
alnFileName = "../../alignmentFiles/mmult3.aux"
workingDirectory = "../../alignmentFiles/"
resultsFileName2 = "../../alignmentFiles/results2.results"
generateMultipleAlignment.getResidues(0.4, resultsFileName2, alnFileName, workingDirectory, False, True, False)
AtomicresultsFileName = "../../alignmentFiles/results2.Atomic"
generateMultipleAlignment.composeAtomicMatrix(resultsFileName2, AtomicresultsFileName)
filteredAtomicMatrixFileName = "../../alignmentFiles/filteredResults2.Atomic"
generateMultipleAlignment.filterAtomicMatrix(AtomicresultsFileName, filteredAtomicMatrixFileName)
caMatrixFileName = "../../alignmentFiles/filteredResults2.Atomic"
dimensions = generateMultipleAlignment.getDimensions(caMatrixFileName)
print "dimensions\n"
print ("%d %d \n" % (dimensions[0], dimensions[1]))
(mask, matrix) = svdAnalysis.loadCAmatrix(caMatrixFileName, dimensions)
(caMatrix, caMask) = svdAnalysis.getCovarianceMatrix(mask, matrix)
print ("matrix.shape %d %d" % (caMatrix.shape[0], caMatrix.shape[1]))
maskFileName = workingDirectory + "mask.atomic"
matrixFileName = workingDirectory + "matrix.atomic"
svdAnalysis.serializeCAMaskAndCAMatrix(mask, matrix, maskFileName, matrixFileName)
realCAMatrixFileName = workingDirectory + "caMatrix.atomic"
realCAMaskFileName = workingDirectory + "caMask.atomic"
svdAnalysis.serializeCAMaskAndCAMatrix(caMask, caMatrix, realCAMaskFileName, realCAMatrixFileName)
def test1loadCAmatrix(self):
"""To generate filteredResults2.Atomic:
1.-generateMultipleAlignment.
alnFileName = "../../alignmentFiles/mmult3.aux"
resultsFileName2 = "../../alignmentFiles/results2.results"
generateMultipleAlignment.getResidues(0.8, resultsFileName2, alnFileName, workingDirectory, False,True, False)
AtomicresultsFileName = "../../alignmentFiles/results2.Atomic"
generateMultipleAlignment.composeAtomicMatrix(resultsFileName2, AtomicresultsFileName)
filteredAtomicMatrixFileName = "../../alignmentFiles/filteredResults2.Atomic"
generateMultipleAlignment.filterAtomicMatrix(AtomicresultsFileName, filteredAtomicMatrixFileName)
"""
caMatrixFileName = "../../alignmentFiles/filteredResults2.Atomic"
dimensions = generateMultipleAlignment.getDimensions(caMatrixFileName)
(mask, cAmatrix) = svdAnalysis.loadCAmatrix(caMatrixFileName, dimensions)
# assert (70,9) == mask.shape
# assert (70,9) == cAmatrix.shape
print ("\n")
print cAmatrix
print ("\n")
print mask
print ("\n")
