def main(input_file):
matrix = parse_file(input_file)
columnmean, coordinates, pc, eigenvalues = pca(matrix)
returns = columnmean, coordinates, pc, eigenvalues
print_top_genes(returns)
f = open("pc_coordinates_" + "learning_topology" + ".out", "w")
for row in coordinates:
row = [str(i) for i in row]
row_to_string = "\t".join(row)
f.write(row_to_string + "\n")
f.close()
def test_pca(self):
if TestCluster.module == 'Bio.Cluster':
from Bio.Cluster import pca
elif TestCluster.module == 'Pycluster':
from Pycluster import pca
data = numpy.array([[3.1, 1.2],
[1.4, 1.3],
[1.1, 1.5],
[2.0, 1.5],
[1.7, 1.9],
[1.7, 1.9],
[5.7, 5.9],
[5.7, 5.9],
[3.1, 3.3],
[5.4, 5.3],
[5.1, 5.5],
[5.0, 5.5],
[5.1, 5.2],
])
mean, coordinates, pc, eigenvalues = pca(data)
self.assertAlmostEqual(mean[0], 3.5461538461538464)
self.assertAlmostEqual(mean[1], 3.5307692307692311)
self.assertAlmostEqual(coordinates[0, 0], 2.0323189722653883)
self.assertAlmostEqual(coordinates[0, 1], 1.2252420399694917)
self.assertAlmostEqual(coordinates[1, 0], 3.0936985166252251)
self.assertAlmostEqual(coordinates[1, 1], -0.10647619705157851)
self.assertAlmostEqual(coordinates[2, 0], 3.1453186907749426)
self.assertAlmostEqual(coordinates[2, 1], -0.46331699855941139)
self.assertAlmostEqual(coordinates[3, 0], 2.5440202962223761)
self.assertAlmostEqual(coordinates[3, 1], 0.20633980959571077)
self.assertAlmostEqual(coordinates[4, 0], 2.4468278463376221)
self.assertAlmostEqual(coordinates[4, 1], -0.28412285736824866)
self.assertAlmostEqual(coordinates[5, 0], 2.4468278463376221)
self.assertAlmostEqual(coordinates[5, 1], -0.28412285736824866)
self.assertAlmostEqual(coordinates[6, 0], -3.2018619434743254)
self.assertAlmostEqual(coordinates[6, 1], 0.019692314198662915)
self.assertAlmostEqual(coordinates[7, 0], -3.2018619434743254)
self.assertAlmostEqual(coordinates[7, 1], 0.019692314198662915)
self.assertAlmostEqual(coordinates[8, 0], 0.46978641990344067)
self.assertAlmostEqual(coordinates[8, 1], -0.17778754731982949)
self.assertAlmostEqual(coordinates[9, 0], -2.5549912731867215)
self.assertAlmostEqual(coordinates[9, 1], 0.19733897451533403)
self.assertAlmostEqual(coordinates[10, 0], -2.5033710990370044)
self.assertAlmostEqual(coordinates[10, 1], -0.15950182699250004)
self.assertAlmostEqual(coordinates[11, 0], -2.4365601663089413)
self.assertAlmostEqual(coordinates[11, 1], -0.23390813900973562)
self.assertAlmostEqual(coordinates[12, 0], -2.2801521629852974)
self.assertAlmostEqual(coordinates[12, 1], 0.0409309711916888)
self.assertAlmostEqual(pc[0, 0], -0.66810932728062988)
self.assertAlmostEqual(pc[0, 1], -0.74406312017235743)
self.assertAlmostEqual(pc[1, 0], 0.74406312017235743)
self.assertAlmostEqual(pc[1, 1], -0.66810932728062988)
self.assertAlmostEqual(eigenvalues[0], 9.3110471246032844)
self.assertAlmostEqual(eigenvalues[1], 1.4437456297481428)
data = numpy.array([[2.3, 4.5, 1.2, 6.7, 5.3, 7.1],
[1.3, 6.5, 2.2, 5.7, 6.2, 9.1],
[3.2, 7.2, 3.2, 7.4, 7.3, 8.9],
[4.2, 5.2, 9.2, 4.4, 6.3, 7.2]])
mean, coordinates, pc, eigenvalues = pca(data)
self.assertAlmostEqual(mean[0], 2.7500)
self.assertAlmostEqual(mean[1], 5.8500)
self.assertAlmostEqual(mean[2], 3.9500)
self.assertAlmostEqual(mean[3], 6.0500)
self.assertAlmostEqual(mean[4], 6.2750)
self.assertAlmostEqual(mean[5], 8.0750)
self.assertAlmostEqual(coordinates[0, 0], 2.6460846688406905)
self.assertAlmostEqual(coordinates[0, 1], -2.1421701432732418)
self.assertAlmostEqual(coordinates[0, 2], -0.56620932754145858)
self.assertAlmostEqual(coordinates[0, 3], 0.0)
self.assertAlmostEqual(coordinates[1, 0], 2.0644120899917544)
self.assertAlmostEqual(coordinates[1, 1], 0.55542108669180323)
self.assertAlmostEqual(coordinates[1, 2], 1.4818772348457117)
self.assertAlmostEqual(coordinates[1, 3], 0.0)
self.assertAlmostEqual(coordinates[2, 0], 1.0686641862092987)
self.assertAlmostEqual(coordinates[2, 1], 1.9994412069101073)
self.assertAlmostEqual(coordinates[2, 2], -1.000720598980291)
self.assertAlmostEqual(coordinates[2, 3], 0.0)
self.assertAlmostEqual(coordinates[3, 0], -5.77916094504174)
self.assertAlmostEqual(coordinates[3, 1], -0.41269215032867046)
self.assertAlmostEqual(coordinates[3, 2], 0.085052691676038017)
self.assertAlmostEqual(coordinates[3, 3], 0.0)
self.assertAlmostEqual(pc[0, 0], -0.26379660005997291)
self.assertAlmostEqual(pc[0, 1], 0.064814972617134495)
self.assertAlmostEqual(pc[0, 2], -0.91763310094893846)
self.assertAlmostEqual(pc[0, 3], 0.26145408875373249)
self.assertAlmostEqual(pc[1, 0], 0.05073770520434398)
self.assertAlmostEqual(pc[1, 1], 0.68616983388698793)
self.assertAlmostEqual(pc[1, 2], 0.13819106187213354)
self.assertAlmostEqual(pc[1, 3], 0.19782544121828985)
self.assertAlmostEqual(pc[2, 0], -0.63000893660095947)
self.assertAlmostEqual(pc[2, 1], 0.091155993862151397)
self.assertAlmostEqual(pc[2, 2], 0.045630391256086845)
self.assertAlmostEqual(pc[2, 3], -0.67456694780914772)
# As the last eigenvalue is zero, the corresponding eigenvector is
# strongly affected by roundoff error, and is not being tested here.
# For PCA, this doesn't matter since all data have a zero coefficient
# along this eigenvector.
self.assertAlmostEqual(eigenvalues[0], 6.7678878332578778)
self.assertAlmostEqual(eigenvalues[1], 3.0108911400291856)
self.assertAlmostEqual(eigenvalues[2], 1.8775592718563467)
self.assertAlmostEqual(eigenvalues[3], 0.0)
from Bio.Cluster import pca
import numpy as np
data = np.array([[1, 2, 3, 4], [5, 6, 7, 8], [9, 10, 11, 12], [0, 1, 2, 3]])
columnmean, coordinates, components, eigenvalues = pca(data)
print("columnmean:", columnmean)
print("coordinates:", coordinates)
print("components:", components)
print("eigenvalues:", eigenvalues)
