def test_calc_initial_position(self):
input = (wordcount.Word(distance=np.array([1, 1, 1, 1])),
wordcount.Word(distance=np.array([1, 1, 1, 0])),
wordcount.Word(distance=np.array([0, 0, 0, 1])),
wordcount.Word(distance=np.array([0, 0, 1, 1])))
got = self.wm.calc_initial_position(input)
dist_zero_one = rogerstanimoto((got[0].x, got[0].y), (got[1].x, got[1].y))
dist_zero_two = rogerstanimoto((got[0].x, got[0].y), (got[2].x, got[2].y))
assert_true(dist_zero_one < dist_zero_two)
def _get_node_distance_matrix(self, datapoint, som_array):
"""Get distance of datapoint and node using Euclidean distance.
Parameters
----------
datapoint : np.array, shape=(X.shape[1])
Datapoint = one row of the dataset `X`
som_array : np.array
Weight vectors of the SOM,
shape = (self.n_rows, self.n_columns, X.shape[1])
Returns
-------
distmat : np.array of float
Distance between datapoint and each SOM node
"""
# algorithms on the full matrix
if self.distance_metric == "euclidean":
return np.linalg.norm(som_array - datapoint, axis=2)
# node-by-node algorithms
distmat = np.zeros((self.n_rows, self.n_columns))
if self.distance_metric == "manhattan":
for node in self.node_list_:
distmat[node] = dist.cityblock(
som_array[node[0], node[1]], datapoint)
elif self.distance_metric == "mahalanobis":
for node in self.node_list_:
som_node = som_array[node[0], node[1]]
cov = np.cov(np.stack((datapoint, som_node), axis=0),
rowvar=False)
cov_pinv = np.linalg.pinv(cov) # pseudo-inverse
distmat[node] = dist.mahalanobis(
datapoint, som_node, cov_pinv)
elif self.distance_metric == "tanimoto":
# Note that this is a binary distance measure.
# Therefore, the vectors have to be converted.
# Source: Melssen 2006, Supervised Kohonen networks for
# classification problems
# VERY SLOW ALGORITHM!!!
threshold = 0.5
for node in self.node_list_:
som_node = som_array[node[0], node[1]]
distmat[node] = dist.rogerstanimoto(
binarize(datapoint.reshape(1, -1), threshold=threshold,
copy=True),
binarize(som_node.reshape(1, -1), threshold=threshold,
copy=True))
elif self.distance_metric == "spectralangle":
for node in self.node_list_:
distmat[node] = np.arccos(np.divide(
np.dot(som_array[node[0], node[1]], datapoint),
np.multiply(np.linalg.norm(som_array),
np.linalg.norm(datapoint))))
return distmat
def calculate_distance_metrics(self, vstrack):
# Subset to top N regions if necessary:
if self.nregions > 0:
self.maxdata = np.maximum(self.maindata, self.vsdata)
self.topidx = np.argpartition(self.maxdata,
-self.nregions)[-self.nregions:]
self.x = self.maindata[self.topidx]
self.y = self.vsdata[self.topidx]
else:
self.x = self.maindata
self.y = self.vsdata
# Calculate normalized euclidean distance:
normdist = 0.5 * (np.var(self.x - self.y) /
(np.var(self.x) + np.var(self.y)))
# Pearson correlation:
cormat = np.corrcoef(self.x, self.y)
cor = cormat[0, 1]
# Spearman correlation:
rho, pval = spearmanr(self.x, self.y)
# Also calculate binary:
if self.ismainobs:
self.bx = 1.0 * (self.x >= 2.0)
else:
self.bx = 1.0 * (self.x >= self.cutoff)
if self.isvsobs:
self.by = 1.0 * (self.y >= 2.0)
else:
self.by = 1.0 * (self.y >= self.cutoff)
# Jaccard distance:
jacc = distance.jaccard(self.bx, self.by)
# Rogers-Tanimoto distance:
rtd = distance.rogerstanimoto(self.bx, self.by)
# Aggregate all:
self.distance_dict[vstrack] = [cor, normdist, rho, jacc, rtd]
print(self.distance_dict[vstrack])
def get_nearest_neighbor(self, x_test, k, sample_class):
distances = []
targets_index = []
for i in range(len(sample_class)):
if (sample_class[i][:] != x_test).any():
if self.distance_calculator == 'jaccard':
distance = dis.jaccard(x_test, sample_class[i][:])
elif self.distance_calculator == 'dice':
distance = dis.dice(x_test, sample_class[i][:])
elif self.distance_calculator == 'correlation':
distance = dis.correlation(x_test, sample_class[i][:])
elif self.distance_calculator == 'yule':
distance = dis.yule(x_test, sample_class[i][:])
elif self.distance_calculator == 'russelo-rao':
distance = dis.russellrao(x_test, sample_class[i][:])
elif self.distance_calculator == 'sokal-michener':
distance = dis.sokalmichener(x_test, sample_class[i][:])
elif self.distance_calculator == 'rogers-tanimoto':
distance = dis.rogerstanimoto(x_test, sample_class[i][:])
elif self.distance_calculator == 'kulzinsky':
distance = dis.kulsinski(x_test, sample_class[i][:])
distances.append([distance, i])
# make a list of the k neighbors' targets
distances.sort()
for i in range(k):
targets_index.append(distances[i][1])
return targets_index
def tanimoto(vector_x: dict, vector_y: dict) -> float:
"""Коэффициент Танимото. Используется для оценки схожести двух образцов.
:param vector_x:
:param vector_y:
:return: [0;+1]
"""
items_vector_x, items_vector_y = Similarity._get_lists(
vector_x, vector_y)
return distance.rogerstanimoto(items_vector_x, items_vector_y)
def rogerstanimoto(self, x=None, y=None, w=None):
"""
田本罗杰斯差异
x = [1, 0, 0]
y = [0, 1, 0]
"""
x = x or self.x
y = y or self.y
w = w or self.w
return distance.rogerstanimoto(x, y, w)
def cross_channel_boolean_distance_features(mask):
"""calculates the cross channel distance features
Calculates the distances across channels
Parameters
----------
mask : 3D array, shape (M, N, C)
The input mask with multiple channels.
Returns
-------
features : dict
dictionary including different distances across channels
"""
features = dict()
for ch1 in range(mask.shape[2]):
for ch2 in range(ch1 + 1, mask.shape[2]):
# rehaping the channels to 1D
channel1 = mask[:, :, ch1].ravel()
channel2 = mask[:, :, ch2].ravel()
# creating the suffix name for better readability
suffix = "_Ch" + str(ch1 + 1) + "_Ch" + str(ch2 + 1)
# storing the distance values
features["dice_distance" + suffix] = dist.dice(channel1, channel2)
features["hamming_distance" + suffix] = dist.hamming(
channel1, channel2)
features["jaccard_distance" + suffix] = dist.jaccard(
channel1, channel2)
features["kulsinski_distance" + suffix] = dist.kulsinski(
channel1, channel2)
features["rogerstanimoto_distance" + suffix] = dist.rogerstanimoto(
channel1, channel2)
features["russellrao_distance" + suffix] = dist.russellrao(
channel1, channel2)
features["sokalmichener_distance" + suffix] = dist.sokalmichener(
channel1, channel2)
features["sokalsneath_distance" + suffix] = dist.sokalsneath(
channel1, channel2)
features["yule_distance" + suffix] = dist.yule(channel1, channel2)
return features
def tanimoto_distances_matrix (param_grid, temporal_filtering, global_nuis_correction, prefs_method, prefs_thr):
from scipy.spatial.distance import rogerstanimoto
wd, xd, dd, labelsf, phenof, dataf, masks, dilmasks, templates, pipe = get_filepaths(temporal_filtering, global_nuis_correction)
#loop will load all classification results dataf (or classify again) and calculate performance metrics
param_gridlist = list(ParameterGrid(param_grid))
n = len(param_gridlist)
tanis = np.zeros((n,n))
ticks = []
for j1idx, j1 in enumerate(param_gridlist):
subjsf = j1['subjsf']
cl = j1['cl']
fidx = dataf.index(subjsf)
maskf = masks[fidx]
result_file1 = os.path.join(wd, get_resultfile_name (subjsf, cl, prefs_method, prefs_thr))
presels_vol1 = get_localizations_from_datashelf(result_file1, subjsf, labelsf, maskf)
ticks.append(subjsf.split('_')[1])
print('Tanimoto comparing ' + result_file1)
for j2idx, j2 in enumerate(param_gridlist):
subjsf = j2['subjsf']
cl = j2['cl']
fidx = dataf.index(subjsf)
maskf = masks[fidx]
result_file2 = os.path.join(wd, get_resultfile_name (subjsf, cl, prefs_method, prefs_thr))
presels_vol2 = get_localizations_from_datashelf(result_file2, subjsf, labelsf, maskf)
print('with ' + result_file2)
tanis[j1idx, j2idx] = rogerstanimoto(presels_vol1.flatten().astype(bool), presels_vol2.flatten().astype(bool))
return tanis, ticks
def calculate_pss(self,
profile,
ignore=None,
method="pairwise"):
"""
Calculate Profiles Similarity Score.
"""
if len(self) != len(profile):
raise ProfileError("Different profiles' lengths")
prof_1 = self
prof_2 = profile
if ignore:
for i in ignore:
try:
prof_1.profile = list(prof_1.profile)
del prof_1.profile[prof_1.query.index(i)]
prof_1.profile = tuple(prof_1.profile)
except IndexError:
raise ProfileError("Element to ignore not in profile")
try:
prof_2.profile = list(prof_2.profile)
del prof_2.profile[prof_2.query.index(i)]
prof_2.profile = tuple(prof_2.profile)
except IndexError:
raise ProfileError("Element to ignore not in profile")
if method == "pairwise":
return sum(a == b for a, b in zip(prof_1.profile, prof_2.profile))
elif method == "jaccard":
return dist.jaccard(prof_1.profile, prof_2.profile)
elif method == "yule":
return dist.yule(prof_1.profile, prof_2.profile)
elif method == "dice":
return dist.dice(prof_1.profile, prof_2.profile)
elif method == "hamming":
return dist.hamming(prof_1.profile, prof_2.profile)
elif method == "kulsinski":
return dist.kulsinski(prof_1.profile, prof_2.profile)
elif method == "rogerstanimoto":
return dist.rogerstanimoto(prof_1.profile, prof_2.profile)
elif method == "russellrao":
return dist.russellrao(prof_1.profile, prof_2.profile)
elif method == "sokalmichener":
return dist.sokalmichener(prof_1.profile, prof_2.profile)
def fitness_func(chromosome):
return 1 - scipy_distance.rogerstanimoto(chromosome, target_chromosome)
def exec_similarity(dct, algorithm):
if validate_similarity_algorithms(dct, algorithm):
return {}
if algorithm == 'braycurtis':
return [
answer.update({
algorithm:
braycurtis(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'canberra':
return [
answer.update({
algorithm:
canberra(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'chebyshev':
return [
answer.update({
algorithm:
chebyshev(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'cityblock':
return [
answer.update({
algorithm:
cityblock(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'correlation':
return [
answer.update({
algorithm:
correlation(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'cosine':
return [
answer.update({
algorithm:
cosine(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'euclidean':
return [
answer.update({
algorithm:
euclidean(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'mahalanobis':
return [
answer.update({
algorithm:
mahalanobis(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
#elif algorithm is 'minkowski':
#return [answer.update({algorithm:minkowski(ndarray_dict(dct['tf_idf']), ndarray_dict(answer['tf_idf']))}) for answer in dct['answers']]
elif algorithm == 'seuclidean':
return [
answer.update({
algorithm:
seuclidean(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'sqeuclidean':
return [
answer.update({
algorithm:
sqeuclidean(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'wminkowski':
return [
answer.update({
algorithm:
wminkowski(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'dice':
return [
answer.update({
algorithm:
dice(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'hamming':
return [
answer.update({
algorithm:
hamming(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'jaccard':
return [
answer.update({
algorithm:
jaccard(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'kulsinski':
return [
answer.update({
algorithm:
kulsinski(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'rogerstanimoto':
return [
answer.update({
algorithm:
rogerstanimoto(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'russellrao':
return [
answer.update({
algorithm:
russellrao(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'sokalmichener':
return [
answer.update({
algorithm:
sokalmichener(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'sokalsneath':
return [
answer.update({
algorithm:
sokalsneath(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
elif algorithm == 'yule':
return [
answer.update({
algorithm:
yule(ndarray_dict(dct['tf_idf']),
ndarray_dict(answer['tf_idf']))
}) for answer in dct['answers']
]
def rogerstanimoto_(x, y):
try:
return rogerstanimoto(x, y)
except ZeroDivisionError:
return 0
desired_organism = sys.argv[5]
except IndexError:
desired_organism = None
if desired_organism is not None:
models = [mod for mod in models if model_info[mod.split(sep)[-1].split('.')[0]][4] == desired_organism]
print ' Predicting for organism : ' + desired_organism
output_name = input_name + '_' + input_name2 + '_out_binary_sim_' + str(threshold) + '_' + desired_organism[:3] + '.txt'
else: output_name = input_name + '_' + input_name2 + '_out_binary_sim_' + str(threshold) + '.txt'
print ' Total Number of Classes : ' + str(len(models))
print ' Using TPR threshold of : ' + str(threshold)
output_name = input_name + '_' + input_name2 + '_out_binary_sim_' + str(threshold) + '.txt'
out_file = open(output_name, 'w')
querymatrix,smiles = importQuery(input_name)
prediction_results = performTargetPrediction(models)
print ' Total Number of Query Molecules file 1 : ' + str(len(querymatrix))
querymatrix,smiles2 = importQuery(input_name2)
prediction_results2 = performTargetPrediction(models)
print ' Total Number of Query Molecules file 2 : ' + str(len(querymatrix))
sim_output = []
sim_output2 = []
for idx in range(prediction_results.shape[1]):
sim_output.append(rogerstanimoto(prediction_results[:,idx],prediction_results2[:,idx]))
sim_output2.append(jaccard(prediction_results[:,idx],prediction_results2[:,idx]))
out_file.write('Compound Pair No.\tSmiles 1\tSmiles 2\tRogers Tanimoto\tJaccard Sim\n')
for idx, comp1 in enumerate(smiles):
comp2 = smiles2[idx]
s = sim_output[idx]
s2 = sim_output2[idx]
out_file.write('\t'.join(map(str,[idx,comp1,comp2,1.0-s,1.0-s2])) + '\n')
print '\n Wrote Results to: ' + output_name
out_file.close()
def rogerstanimoto_(x, y):
try:
return rogerstanimoto(x, y)
except ZeroDivisionError:
return 0
def distance(self, vector1, vector2, type_):
"""
Calculate distance between two vectors.
Args:
vector1 (list of int/float/bool): Vector in vector space
vector2 (list of int/float/bool): Vector in vector space
type_ (str): Type of distance calculation. Allowed types are:
* For numeric vectors *
- braycurtis: Computes the Bray-Curtis distance between two arrays.
- canberra: Computes the Canberra distance between two arrays.
- chebyshev: Computes the Chebyshev distance.
- cityblock: Computes the City Block (Manhattan) distance.
- correlation: Computes the correlation distance between two arrays.
- cosine: Computes the Cosine distance between arrays.
- euclidean: Computes the Euclidean distance between two arrays.
- sqeuclidean: Computes the squared Euclidean distance between two arrays.
* For boolean vectors *
- dice: Computes the Dice dissimilarity between two boolean arrays.
- hamming: Computes the Hamming distance between two arrays.
- jaccard: Computes the Jaccard-Needham dissimilarity between two boolean arrays.
- kulsinski: Computes the Kulsinski dissimilarity between two boolean arrays.
- rogerstanimoto: Computes the Rogers-Tanimoto dissimilarity between two boolean arrays.
- russellrao: Computes the Russell-Rao dissimilarity between two boolean arrays.
- sokalmichener: Computes the Sokal-Michener dissimilarity between two boolean arrays.
- sokalsneath: Computes the Sokal-Sneath dissimilarity between two boolean arrays.
- yule: Computes the Yule dissimilarity between two boolean arrays.
Returns:
float: Distance between vectors.
"""
if type_ == "braycurtis":
return distance.braycurtis(vector1, vector2)
elif type_ == "canberra":
return distance.canberra(vector1, vector2)
elif type_ == "chebyshev":
return distance.chebyshev(vector1, vector2)
elif type_ == "cityblock":
return distance.cityblock(vector1, vector2)
elif type_ == "correlation":
return distance.correlation(vector1, vector2)
elif type_ == "cosine":
return distance.cosine(vector1, vector2)
elif type_ == "euclidean":
return distance.euclidean(vector1, vector2)
elif type_ == "sqeuclidean":
return distance.sqeuclidean(vector1, vector2)
elif type_ == "dice":
return distance.dice(vector1, vector2)
elif type_ == "hamming":
return distance.hamming(vector1, vector2)
elif type_ == "jaccard":
return distance.jaccard(vector1, vector2)
elif type_ == "kulsinski":
return distance.kulsinski(vector1, vector2)
elif type_ == "kulsinski":
return distance.kulsinski(vector1, vector2)
elif type_ == "rogerstanimoto":
return distance.rogerstanimoto(vector1, vector2)
elif type_ == "russellrao":
return distance.russellrao(vector1, vector2)
elif type_ == "sokalmichener":
return distance.sokalmichener(vector1, vector2)
elif type_ == "sokalsneath":
return distance.sokalsneath(vector1, vector2)
elif type_ == "yule":
return distance.yule(vector1, vector2)
else:
raise ValueError(
"""Wrong value for type_. Please enter one of supported values.
Type help(distance) to see supported values.""")
def tanimoto_similarity(base, target):
evaluation = rogerstanimoto(base, target)
return evaluation
def getDistancesFromMeds(noBits, signBitOption, aggregateTreatmentsOption,
binaryDistanceMethod, useManualDistanceMethod,
segmentLength):
# Pull in previously made med sequence matrix and delete the patient identifiers
treatmentData = np.delete(gv.medSequenceMatrix, 0, 1)
# Create an array of the binary strings which are encoded with each segment's treatments
monthlyBinaryTreatmentVectors = []
for i in range(0, treatmentData.shape[0]):
temp = []
for j in range(0, segmentLength):
binaryTreatment = convTreatmentToBinaryArray(
treatmentData[i, j], signBitOption, noBits)
temp.append(binaryTreatment)
monthlyBinaryTreatmentVectors.append(temp)
monthlyBinaryTreatmentVectors = np.array(monthlyBinaryTreatmentVectors)
binaryTreatmentVectors = []
# If the aggregate option is chosen, then logically OR each binary string in a segment to form one
# binary string of length noBits per each segment
if (aggregateTreatmentsOption):
boolMonthlyTreatVectors = monthlyBinaryTreatmentVectors.astype(bool)
for i in range(monthlyBinaryTreatmentVectors.shape[0]):
rowOfMonths = np.zeros(noBits + signBitOption).astype(bool)
for j in range(monthlyBinaryTreatmentVectors.shape[1]):
rowOfMonths = np.logical_or(rowOfMonths,
boolMonthlyTreatVectors[i, j])
binaryTreatmentVectors.append(rowOfMonths.astype(int))
# If the aggregate treatment option is not chosen, then monthly granujlarity is preserved by
# Concatenating every binary string in a segment to form one long binary string of length
# noBits * segmentLength per each segment
else:
for i in range(monthlyBinaryTreatmentVectors.shape[0]):
rowOfMonths = np.array([])
for j in range(monthlyBinaryTreatmentVectors.shape[1]):
rowOfMonths = np.concatenate(
[rowOfMonths, monthlyBinaryTreatmentVectors[i, j]])
binaryTreatmentVectors.append(rowOfMonths)
binaryTreatmentVectors = np.array(binaryTreatmentVectors)
# Initialize distance matrix and compute distances based on the selected distance method
distanceMatrix = np.zeros(
(len(binaryTreatmentVectors), len(binaryTreatmentVectors)))
if binaryDistanceMethod == 1: #Sokal Michener
for i in range(binaryTreatmentVectors.shape[0]):
for j in range(i, binaryTreatmentVectors.shape[0]):
if useManualDistanceMethod:
dist = np.sum(
np.logical_xor(binaryTreatmentVectors[i].astype(bool),
binaryTreatmentVectors[j].astype(
bool))) / (noBits + signBitOption)
else:
dist = distance.hamming(
binaryTreatmentVectors[i].astype(bool),
binaryTreatmentVectors[j].astype(bool))
distanceMatrix[i, j] = dist
distanceMatrix[j, i] = dist
elif binaryDistanceMethod == 2: #Jaccard
for i in range(binaryTreatmentVectors.shape[0]):
for j in range(i, binaryTreatmentVectors.shape[0]):
if np.array_equal(binaryTreatmentVectors[i],
binaryTreatmentVectors[j]):
dist = 0
else:
if useManualDistanceMethod:
numerator = np.sum(
np.logical_xor(
binaryTreatmentVectors[i].astype(bool),
binaryTreatmentVectors[j].astype(bool)))
denomenator = np.sum(
np.logical_or(
binaryTreatmentVectors[i].astype(bool),
binaryTreatmentVectors[j].astype(bool)))
dist = numerator / denomenator
else:
dist = distance.jaccard(
binaryTreatmentVectors[i].astype(bool),
binaryTreatmentVectors[j].astype(bool))
distanceMatrix[i, j] = dist
distanceMatrix[j, i] = dist
elif binaryDistanceMethod == 3: #Rogers Tanimoto
for i in range(binaryTreatmentVectors.shape[0]):
for j in range(i, binaryTreatmentVectors.shape[0]):
if useManualDistanceMethod:
numerator = 2 * np.sum(
np.logical_xor(binaryTreatmentVectors[i].astype(bool),
binaryTreatmentVectors[j].astype(bool)))
denomenator = np.sum(
np.logical_xor(binaryTreatmentVectors[i].astype(bool),
binaryTreatmentVectors[j].astype(
bool))) + noBits + signBitOption
dist = numerator / denomenator
else:
dist = distance.rogerstanimoto(
binaryTreatmentVectors[i].astype(bool),
binaryTreatmentVectors[j].astype(bool))
distanceMatrix[i, j] = dist
distanceMatrix[j, i] = dist
else: #Sokal Sneath II
for i in range(binaryTreatmentVectors.shape[0]):
for j in range(binaryTreatmentVectors.shape[0]):
if np.array_equal(binaryTreatmentVectors[i],
binaryTreatmentVectors[j]):
dist = 0
else:
if useManualDistanceMethod:
numerator = 2 * np.sum(
np.logical_xor(
binaryTreatmentVectors[i].astype(bool),
binaryTreatmentVectors[j].astype(bool)))
denomenator = (2 * np.sum(
np.logical_xor(
binaryTreatmentVectors[i].astype(bool),
binaryTreatmentVectors[j].astype(bool))
)) + np.sum(
np.logical_and(
binaryTreatmentVectors[i].astype(bool),
binaryTreatmentVectors[j].astype(bool)))
dist = numerator / denomenator
else:
dist = distance.sokalsneath(
binaryTreatmentVectors[i].astype(bool),
binaryTreatmentVectors[j].astype(bool))
distanceMatrix[i, j] = dist
# Normalize distance matrix by dividing the whole thing by the largest value.
max = np.amax(distanceMatrix)
distanceMatrix = np.divide(distanceMatrix, max)
return distanceMatrix
def my_dist(u, v):
return cosine(u, v) * yule(u, v) * braycurtis(u, v) * np.abs(
rogerstanimoto(u, v))
def tanimoto_dist(A, B):
return rogerstanimoto(A, B)
jaccard_sim = np.zeros((3, 3))
for i in range(0, 3):
for j in range(0, 3):
if i == j:
jaccard_sim[i, j] = 1
else:
jaccard_sim[i, j] = jaccard_score(y[i, :], y[j, :])
jaccard_sim
#simple matching score
simatch_sim = np.zeros((3, 3))
for i in range(0, 3):
for j in range(0, 3):
if i == j:
simatch_sim[i, j] = 1
else:
simatch_sim[i, j] = sum(y[i, :] == y[j, :]) / 8
simatch_sim
#roger score
rogerstan_sim = np.zeros((3, 3))
for i in range(0, 3):
for j in range(0, 3):
if i == j:
rogerstan_sim[i, j] = 1
else:
rogerstan_sim[i, j] = 1 - distance.rogerstanimoto(y[i, :], y[j, :])
rogerstan_sim
def rogerstanimoto(app1SyscallsVector, app2SyscallsVector):
return spDist.rogerstanimoto(app1SyscallsVector, app2SyscallsVector)
import sys
import numpy as np
from scipy.spatial import distance
filename1 = sys.argv[1]
filename2 = sys.argv[2]
pad1 = np.loadtxt(filename1,
dtype='float',
delimiter='\t',
usecols=(5),
unpack=True,
skiprows=11)
pad2 = np.loadtxt(filename2,
dtype='float',
delimiter='\t',
usecols=(5),
unpack=True,
skiprows=11)
dE = distance.euclidean(pad1, pad2)
rog = distance.rogerstanimoto(pad1, pad2)
print("Euclidean", dE)
print("Rogers", rog)
print ' Total Number of Classes : ' + str(len(models))
print ' Using TPR threshold of : ' + str(threshold)
output_name = input_name + '_' + input_name2 + '_out_binary_sim_' + str(
threshold) + '.txt'
out_file = open(output_name, 'w')
querymatrix, smiles, ids = importQuery(input_name)
prediction_results = performTargetPrediction(models)
print ' Total Number of Query Molecules file 1 : ' + str(len(querymatrix))
querymatrix, smiles2, ids2 = importQuery(input_name2)
prediction_results2 = performTargetPrediction(models)
print ' Total Number of Query Molecules file 2 : ' + str(len(querymatrix))
sim_output = []
sim_output2 = []
for idx in range(prediction_results.shape[1]):
sim_output.append(
rogerstanimoto(prediction_results[:, idx],
prediction_results2[:, idx]))
sim_output2.append(
jaccard(prediction_results[:, idx], prediction_results2[:, idx]))
out_file.write(
'Compound Pair No.\tSmiles 1\tSmiles 2\tRogers Tanimoto\tJaccard Sim\n'
)
for idx, comp1 in enumerate(ids):
comp2 = ids2[idx]
s = sim_output[idx]
s2 = sim_output2[idx]
out_file.write(
'\t'.join(map(str, [idx, comp1, comp2, 1.0 - s, 1.0 - s2])) + '\n')
print '\n Wrote Results to: ' + output_name
out_file.close()
def do_roger_tanimoto(m, rogtan, vec):
for i in range(m):
for j in range(m):
rogtan[i, j] = distance.rogerstanimoto(vec[i], vec[j])
return rogtan
def main():
from scipy.spatial import distance
a = np.array([1, 2, 43])
b = np.array([3, 2, 1])
d = Distance()
print('-----------------------------------------------------------------')
print('My braycurtis: {}'.format(d.braycurtis(a, b)))
print('SciPy braycurtis: {}'.format(distance.braycurtis(a, b)))
print('-----------------------------------------------------------------')
print('My canberra: {}'.format(d.canberra(a, b)))
print('SciPy canberra: {}'.format(distance.canberra(a, b)))
print('-----------------------------------------------------------------')
print('My chebyshev: {}'.format(d.chebyshev(a, b)))
print('SciPy chebyshev: {}'.format(distance.chebyshev(a, b)))
print('-----------------------------------------------------------------')
print('My cityblock: {}'.format(d.cityblock(a, b)))
print('SciPy cityblock: {}'.format(distance.cityblock(a, b)))
print('-----------------------------------------------------------------')
print('My correlation: {}'.format(d.correlation(a, b)))
print('SciPy correlation: {}'.format(distance.correlation(a, b)))
print('-----------------------------------------------------------------')
print('My euclidean: {}'.format(d.euclidean(a, b)))
print('SciPy euclidean: {}'.format(distance.euclidean(a, b)))
print('-----------------------------------------------------------------')
print('My hamming: {}'.format(d.hamming(a, b)))
print('SciPy hamming: {}'.format(distance.hamming(a, b)))
print('-----------------------------------------------------------------')
print('My jaccard: {}'.format(d.jaccard(a, b)))
print('SciPy jaccard: {}'.format(distance.jaccard(a, b)))
print('-----------------------------------------------------------------')
print('My manhattan: {}'.format(d.cityblock(a, b)))
print('SciPy manhattan: {}'.format(distance.cityblock(a, b)))
print('-----------------------------------------------------------------')
print('My cosine: {}'.format(d.cosine(a, b)))
print('SciPy cosine: {}'.format(distance.cosine(a, b)))
print('-----------------------------------------------------------------')
print('My dice: {}'.format(d.dice(a, b)))
print('SciPy dice: {}'.format(distance.dice(a, b)))
print('-----------------------------------------------------------------')
print('My kulsinski: {}'.format(d.kulsinski(a, b)))
print('SciPy kulsinski: {}'.format(distance.kulsinski(a, b)))
print('-----------------------------------------------------------------')
iv = np.array([[1, 0.5, 0.5], [0.5, 1, 0.5], [0.5, 0.5, 1]])
print('My mahalanobis: {}'.format(d.mahalanobis(a, b, iv)))
print('SciPy mahalanobis: {}'.format(distance.mahalanobis(a, b, iv)))
print('-----------------------------------------------------------------')
print('My seuclidean: {}'.format(
d.seuclidean(a, b, np.array([0.1, 0.1, 0.1]))))
print('SciPy seuclidean: {}'.format(
distance.seuclidean(a, b, [0.1, 0.1, 0.1])))
print('-----------------------------------------------------------------')
print('My sokalmichener: {}'.format(d.sokalmichener(a, b)))
print('SciPy sokalmichener: {}'.format(distance.sokalmichener(a, b)))
print('-----------------------------------------------------------------')
print('My sokal_sneath: {}'.format(d.sokalsneath(a, b)))
print('SciPy sokal_sneath: {}'.format(distance.sokalsneath(a, b)))
print('-----------------------------------------------------------------')
print('My sqeuclidean: {}'.format(d.sqeuclidean(a, b)))
print('SciPy sqeuclidean: {}'.format(distance.sqeuclidean(a, b)))
print('-----------------------------------------------------------------')
print('My minkowski: {}'.format(d.minkowski(a, b, 2)))
print('SciPy minkowski: {}'.format(distance.minkowski(a, b, 2)))
print('-----------------------------------------------------------------')
print('My rogerstanimoto: {}'.format(d.rogerstanimoto(a, b)))
print('SciPy rogerstanimoto: {}'.format(distance.rogerstanimoto(a, b)))
print('-----------------------------------------------------------------')
print('My russellrao: {}'.format(d.russellrao(a, b)))
print('SciPy russellrao: {}'.format(distance.russellrao(a, b)))
print('-----------------------------------------------------------------')
print('My wminkowski: {}'.format(d.wminkowski(a, b, 2, np.ones(3))))
print('SciPy wminkowski: {}'.format(
distance.wminkowski(a, b, 2, np.ones(3))))
print('-----------------------------------------------------------------')
print('My yule: {}'.format(d.yule(a, b)))
print('SciPy yule: {}'.format(distance.yule(a, b)))
print('-----------------------------------------------------------------')