def test_summarize_pcoas(self):
"""summarize_pcoas works
"""
master_pcoa = [['1', '2', '3'], \
array([[-1.0, 0.0, 1.0], [2.0, 4.0, -4.0]]), \
array([.76, .24])]
jn1 = [['1', '2', '3'], \
array([[1.2, 0.1, -1.2],[-2.5, -4.0, 4.5]]), \
array([0.80, .20])]
jn2 = [['1', '2', '3'], \
array([[-1.4, 0.05, 1.3],[2.6, 4.1, -4.7]]), \
array([0.76, .24])]
jn3 = [['1', '2', '3'], \
array([[-1.5, 0.05, 1.6],[2.4, 4.0, -4.8]]), \
array([0.84, .16])]
jn4 = [['1', '2', '3'], \
array([[-1.5, 0.05, 1.6],[2.4, 4.0, -4.8]]), \
array([0.84, .16])]
support_pcoas = [jn1, jn2, jn3, jn4]
#test with the ideal_fourths option
matrix_average, matrix_low, matrix_high, eigval_average, m_names = \
summarize_pcoas(master_pcoa, support_pcoas, 'ideal_fourths',
apply_procrustes=False)
self.assertEqual(m_names, ['1', '2', '3'])
self.assertFloatEqual(matrix_average[(0, 0)], -1.4)
self.assertFloatEqual(matrix_average[(0, 1)], 0.0125)
self.assertFloatEqual(matrix_low[(0, 0)], -1.5)
self.assertFloatEqual(matrix_high[(0, 0)], -1.28333333)
self.assertFloatEqual(matrix_low[(0, 1)], -0.0375)
self.assertFloatEqual(matrix_high[(0, 1)], 0.05)
self.assertFloatEqual(eigval_average[0], 0.81)
self.assertFloatEqual(eigval_average[1], 0.19)
#test with the IQR option
matrix_average, matrix_low, matrix_high, eigval_average, m_names = \
summarize_pcoas(master_pcoa, support_pcoas, method='IQR',
apply_procrustes=False)
self.assertFloatEqual(matrix_low[(0, 0)], -1.5)
self.assertFloatEqual(matrix_high[(0, 0)], -1.3)
#test with procrustes option followed by sdev
m, m1, msq = procrustes(master_pcoa[1], jn1[1])
m, m2, msq = procrustes(master_pcoa[1], jn2[1])
m, m3, msq = procrustes(master_pcoa[1], jn3[1])
m, m4, msq = procrustes(master_pcoa[1], jn4[1])
matrix_average, matrix_low, matrix_high, eigval_average, m_names = \
summarize_pcoas(master_pcoa, support_pcoas, method='sdev',
apply_procrustes=True)
x = array([m1[0, 0], m2[0, 0], m3[0, 0], m4[0, 0]])
self.assertEqual(x.mean(), matrix_average[0, 0])
self.assertEqual(-x.std(ddof=1) / 2, matrix_low[0, 0])
self.assertEqual(x.std(ddof=1) / 2, matrix_high[0, 0])
def test_summarize_pcoas(self):
"""summarize_pcoas works
"""
master_pcoa = [['1', '2', '3'], \
array([[-1.0, 0.0, 1.0], [2.0, 4.0, -4.0]]), \
array([.76, .24])]
jn1 = [['1', '2', '3'], \
array([[1.2, 0.1, -1.2],[-2.5, -4.0, 4.5]]), \
array([0.80, .20])]
jn2 = [['1', '2', '3'], \
array([[-1.4, 0.05, 1.3],[2.6, 4.1, -4.7]]), \
array([0.76, .24])]
jn3 = [['1', '2', '3'], \
array([[-1.5, 0.05, 1.6],[2.4, 4.0, -4.8]]), \
array([0.84, .16])]
jn4 = [['1', '2', '3'], \
array([[-1.5, 0.05, 1.6],[2.4, 4.0, -4.8]]), \
array([0.84, .16])]
support_pcoas = [jn1, jn2, jn3, jn4]
#test with the ideal_fourths option
matrix_average, matrix_low, matrix_high, eigval_average, m_names = \
summarize_pcoas(master_pcoa, support_pcoas, 'ideal_fourths',
apply_procrustes=False)
self.assertEqual(m_names, ['1', '2', '3'])
self.assertFloatEqual(matrix_average[(0,0)], -1.4)
self.assertFloatEqual(matrix_average[(0,1)], 0.0125)
self.assertFloatEqual(matrix_low[(0,0)], -1.5)
self.assertFloatEqual(matrix_high[(0,0)], -1.28333333)
self.assertFloatEqual(matrix_low[(0,1)], -0.0375)
self.assertFloatEqual(matrix_high[(0,1)], 0.05)
self.assertFloatEqual(eigval_average[0], 0.81)
self.assertFloatEqual(eigval_average[1], 0.19)
#test with the IQR option
matrix_average, matrix_low, matrix_high, eigval_average, m_names = \
summarize_pcoas(master_pcoa, support_pcoas, method='IQR',
apply_procrustes=False)
self.assertFloatEqual(matrix_low[(0,0)], -1.5)
self.assertFloatEqual(matrix_high[(0,0)], -1.3)
#test with procrustes option followed by sdev
m, m1, msq = procrustes(master_pcoa[1],jn1[1])
m, m2, msq = procrustes(master_pcoa[1],jn2[1])
m, m3, msq = procrustes(master_pcoa[1],jn3[1])
m, m4, msq = procrustes(master_pcoa[1],jn4[1])
matrix_average, matrix_low, matrix_high, eigval_average, m_names = \
summarize_pcoas(master_pcoa, support_pcoas, method='sdev',
apply_procrustes=True)
x = array([m1[0,0],m2[0,0],m3[0,0],m4[0,0]])
self.assertEqual(x.mean(),matrix_average[0,0])
self.assertEqual(-x.std(ddof=1)/2,matrix_low[0,0])
self.assertEqual(x.std(ddof=1)/2,matrix_high[0,0])
def get_coord(coord_fname, method="IQR"):
"""Opens and returns coords location matrix and metadata.
Also two spread matrices (+/-) if passed a dir of coord files.
If only a single coord file, spread matrices are returned as None.
"""
if not os.path.isdir(coord_fname):
try:
coord_f = open(coord_fname, 'U')
except (TypeError, IOError):
raise MissingFileError('Coord file required for this analysis')
coord_header, coords, eigvals, pct_var = parse_coords(coord_f)
return [coord_header, coords, eigvals, pct_var, None, None]
else:
master_pcoa, support_pcoas = load_pcoa_files(coord_fname)
# get Summary statistics
coords, coords_low, coords_high, eigval_average, coord_header = \
summarize_pcoas(master_pcoa, support_pcoas, method=method)
pct_var = master_pcoa[3] # should be getting this from an average
# make_3d_plots expects coord_header to be a python list
coord_header = list(master_pcoa[0])
return ([
coord_header, coords, eigval_average, pct_var, coords_low,
coords_high
])
def get_coord(coord_fname, method="IQR"):
"""Opens and returns coords location matrix and metadata.
Also two spread matrices (+/-) if passed a dir of coord files.
If only a single coord file, spread matrices are returned as None.
"""
if not os.path.isdir(coord_fname):
try:
coord_f = open(coord_fname, 'U')
except (TypeError, IOError):
raise MissingFileError('Coord file required for this analysis')
coord_header, coords, eigvals, pct_var = parse_coords(coord_f)
return [coord_header, coords, eigvals, pct_var, None, None]
else:
master_pcoa, support_pcoas = load_pcoa_files(coord_fname)
# get Summary statistics
coords, coords_low, coords_high, eigval_average, coord_header = \
summarize_pcoas(master_pcoa, support_pcoas, method=method)
pct_var = master_pcoa[3] # should be getting this from an average
# make_3d_plots expects coord_header to be a python list
coord_header = list(master_pcoa[0])
return (
[coord_header,
coords,
eigval_average,
pct_var,
coords_low,
coords_high]
)
def generate_2d_plots(prefs,data,html_dir_path,data_dir_path,filename,
background_color,label_color,generate_scree):
"""Generate interactive 2D scatterplots"""
coord_tups = [("1", "2"), ("3", "2"), ("1", "3")]
mapping=data['map']
out_table=''
#Iterate through prefs and generate html files for each colorby option
#Sort by the column name first
sample_location={}
groups_and_colors=iter_color_groups(mapping,prefs)
groups_and_colors=list(groups_and_colors)
radiobuttons = []
for i in range(len(groups_and_colors)):
labelname=groups_and_colors[i][0] #'EnvoID'
groups=groups_and_colors[i][1] #defaultdict(<type 'list'>, {'mangrove biome/marine habitat/ocean water': ['BBA.number1.filt..660397', 'BBA.number2.filt..660380', ...}
colors=groups_and_colors[i][2] #{'mangrove biome/marine habitat/ocean water': 'red5', 'Small lake biome/marine habitat/saline lake sediment': 'cyan1',
data_colors=groups_and_colors[i][3]#{'orange1': <qiime.colors.Color object at 0x25f1210>, 'orange3':
data_color_order=groups_and_colors[i][4]#['red1', 'blue1', 'orange1', 'green1', 'purple1', 'yellow1', 'cyan1', 'pink1', 'teal1', ...]
data_file_dir_path = get_random_directory_name(output_dir=data_dir_path)
new_link=os.path.split(data_file_dir_path)
data_file_link=os.path.join('.', os.path.split(new_link[-2])[-1], \
new_link[-1])
new_col_name=labelname
img_data = {}
plot_label=labelname
if data.has_key('support_pcoas'):
matrix_average, matrix_low, matrix_high, eigval_average, m_names = \
summarize_pcoas(data['coord'], data['support_pcoas'],
method=data['ellipsoid_method'])
data['coord'] = \
(m_names,matrix_average,data['coord'][2],data['coord'][3])
for i in range(len(m_names)):
sample_location[m_names[i]]=i
else:
matrix_average = None
matrix_low = None
matrix_high = None
eigval_average = None
m_names = None
iterator=0
for coord_tup in coord_tups: # change, if you want more thatn one PCoA plot! (i.e involving PC3)
if isarray(matrix_low) and isarray(matrix_high) and \
isarray(matrix_average):
coord_1r=asarray(matrix_low)
coord_2r=asarray(matrix_high)
mat_ave=asarray(matrix_average)
else:
coord_1r=None
coord_2r=None
mat_ave=None
sample_location=None
coord_1, coord_2 = coord_tup
img_data[coord_tup] = draw_pcoa_graph(plot_label,data_file_dir_path,
data_file_link,coord_1,coord_2,
coord_1r, coord_2r, mat_ave,\
sample_location,
data,prefs,groups,colors,
background_color,label_color,
data_colors,data_color_order,
generate_eps=True)
radiobuttons.append(RADIO % (data_file_link, labelname))
if i == 0: ## only create first table!
out_table += TABLE_HTML % (labelname,
"<br>".join(img_data[("1", "2")]),
"<br>".join(img_data[("3", "2")]),
"<br>".join(img_data[("1", "3")]))
if generate_scree:
data_file_dir_path = get_random_directory_name(output_dir = data_dir_path)
new_link = os.path.split(data_file_dir_path)
data_file_link = os.path.join('.', os.path.split(new_link[-2])[-1], new_link[-1])
img_src, download_link = draw_scree_graph(data_file_dir_path, data_file_link, background_color,
label_color, generate_eps = True, data = data)
out_table += SCREE_TABLE_HTML % ("<br>".join((img_src, download_link)))
out_table = "\n".join(radiobuttons) + out_table
outfile = create_html_filename(filename,'.html')
outfile = os.path.join(html_dir_path,outfile)
write_html_file(out_table,outfile)
def generate_2d_plots(prefs, data, html_dir_path, data_dir_path, filename,
background_color, label_color, generate_scree):
"""Generate interactive 2D scatterplots"""
coord_tups = [("1", "2"), ("3", "2"), ("1", "3")]
mapping = data['map']
out_table = ''
#Iterate through prefs and generate html files for each colorby option
#Sort by the column name first
sample_location = {}
groups_and_colors = iter_color_groups(mapping, prefs)
groups_and_colors = list(groups_and_colors)
for i in range(len(groups_and_colors)):
labelname = groups_and_colors[i][0]
groups = groups_and_colors[i][1]
colors = groups_and_colors[i][2]
data_colors = groups_and_colors[i][3]
data_color_order = groups_and_colors[i][4]
data_file_dir_path = get_random_directory_name(
output_dir=data_dir_path)
new_link = os.path.split(data_file_dir_path)
data_file_link=os.path.join('.', os.path.split(new_link[-2])[-1], \
new_link[-1])
new_col_name = labelname
img_data = {}
plot_label = labelname
if data.has_key('support_pcoas'):
matrix_average, matrix_low, matrix_high, eigval_average, m_names = \
summarize_pcoas(data['coord'], data['support_pcoas'],
method=data['ellipsoid_method'])
data['coord'] = \
(m_names,matrix_average,data['coord'][2],data['coord'][3])
for i in range(len(m_names)):
sample_location[m_names[i]] = i
else:
matrix_average = None
matrix_low = None
matrix_high = None
eigval_average = None
m_names = None
iterator = 0
for coord_tup in coord_tups:
if isarray(matrix_low) and isarray(matrix_high) and \
isarray(matrix_average):
coord_1r = asarray(matrix_low)
coord_2r = asarray(matrix_high)
mat_ave = asarray(matrix_average)
else:
coord_1r = None
coord_2r = None
mat_ave = None
sample_location = None
coord_1, coord_2 = coord_tup
img_data[coord_tup] = draw_pcoa_graph(plot_label,data_file_dir_path,
data_file_link,coord_1,coord_2,
coord_1r, coord_2r, mat_ave,\
sample_location,
data,prefs,groups,colors,
background_color,label_color,
data_colors,data_color_order,
generate_eps=True)
out_table += TABLE_HTML % (labelname, "<br>".join(
img_data[("1", "2")]), "<br>".join(
img_data[("3", "2")]), "<br>".join(img_data[("1", "3")]))
if generate_scree:
data_file_dir_path = get_random_directory_name(
output_dir=data_dir_path)
new_link = os.path.split(data_file_dir_path)
data_file_link = os.path.join('.',
os.path.split(new_link[-2])[-1],
new_link[-1])
img_src, download_link = draw_scree_graph(data_file_dir_path,
data_file_link,
background_color,
label_color,
generate_eps=True,
data=data)
out_table += SCREE_TABLE_HTML % ("<br>".join((img_src, download_link)))
outfile = create_html_filename(filename, '.html')
outfile = os.path.join(html_dir_path, outfile)
write_html_file(out_table, outfile)
def generate_2d_plots(prefs, data, html_dir_path, data_dir_path, filename,
background_color, label_color, generate_scree):
"""Generate interactive 2D scatterplots"""
coord_tups = [("1", "2"), ("3", "2"), ("1", "3")]
mapping = data['map']
out_table = ''
# Iterate through prefs and generate html files for each colorby option
# Sort by the column name first
sample_location = {}
groups_and_colors = iter_color_groups(mapping, prefs)
groups_and_colors = list(groups_and_colors)
for i in range(len(groups_and_colors)):
labelname = groups_and_colors[i][0]
groups = groups_and_colors[i][1]
colors = groups_and_colors[i][2]
data_colors = groups_and_colors[i][3]
data_color_order = groups_and_colors[i][4]
data_file_dir_path = mkdtemp(dir=data_dir_path)
new_link = os.path.split(data_file_dir_path)
data_file_link = os.path.join('.', os.path.split(new_link[-2])[-1],
new_link[-1])
new_col_name = labelname
img_data = {}
plot_label = labelname
if 'support_pcoas' in data:
matrix_average, matrix_low, matrix_high, eigval_average, m_names = \
summarize_pcoas(data['coord'], data['support_pcoas'],
method=data['ellipsoid_method'])
data['coord'] = \
(m_names, matrix_average, data['coord'][2], data['coord'][3])
for i in range(len(m_names)):
sample_location[m_names[i]] = i
else:
matrix_average = None
matrix_low = None
matrix_high = None
eigval_average = None
m_names = None
iterator = 0
for coord_tup in coord_tups:
if isarray(matrix_low) and isarray(matrix_high) and \
isarray(matrix_average):
coord_1r = asarray(matrix_low)
coord_2r = asarray(matrix_high)
mat_ave = asarray(matrix_average)
else:
coord_1r = None
coord_2r = None
mat_ave = None
sample_location = None
coord_1, coord_2 = coord_tup
img_data[coord_tup] = draw_pcoa_graph(
plot_label, data_file_dir_path,
data_file_link, coord_1, coord_2,
coord_1r, coord_2r, mat_ave,
sample_location,
data, prefs, groups, colors,
background_color, label_color,
data_colors, data_color_order,
generate_eps=True)
out_table += TABLE_HTML % (labelname,
"<br>".join(img_data[("1", "2")]),
"<br>".join(img_data[("3", "2")]),
"<br>".join(img_data[("1", "3")]))
if generate_scree:
data_file_dir_path = mkdtemp(dir=data_dir_path)
new_link = os.path.split(data_file_dir_path)
data_file_link = os.path.join(
'.',
os.path.split(new_link[-2])[-1],
new_link[-1])
img_src, download_link = draw_scree_graph(
data_file_dir_path, data_file_link, background_color,
label_color, generate_eps=True, data=data)
out_table += SCREE_TABLE_HTML % ("<br>".join((img_src, download_link)))
outfile = create_html_filename(filename, '.html')
outfile = os.path.join(html_dir_path, outfile)
write_html_file(out_table, outfile)
def generate_2d_plots(prefs,data,html_dir_path,data_dir_path,filename,
background_color,label_color,generate_scree):
"""Generate interactive 2D scatterplots"""
coord_tups = [("1", "2"), ("3", "2"), ("1", "3")]
mapping=data['map']
out_table=''
#Iterate through prefs and generate html files for each colorby option
#Sort by the column name first
sample_location={}
groups_and_colors=iter_color_groups(mapping,prefs)
groups_and_colors=list(groups_and_colors)
radiobuttons = []
for i in range(len(groups_and_colors)):
labelname=groups_and_colors[i][0] #'EnvoID'
groups=groups_and_colors[i][1] #defaultdict(<type 'list'>, {'mangrove biome/marine habitat/ocean water': ['BBA.number1.filt..660397', 'BBA.number2.filt..660380', ...}
colors=groups_and_colors[i][2] #{'mangrove biome/marine habitat/ocean water': 'red5', 'Small lake biome/marine habitat/saline lake sediment': 'cyan1',
data_colors=groups_and_colors[i][3]#{'orange1': <qiime.colors.Color object at 0x25f1210>, 'orange3':
data_color_order=groups_and_colors[i][4]#['red1', 'blue1', 'orange1', 'green1', 'purple1', 'yellow1', 'cyan1', 'pink1', 'teal1', ...]
data_file_dir_path = get_random_directory_name(output_dir=data_dir_path)
new_link=os.path.split(data_file_dir_path)
data_file_link=os.path.join('.', os.path.split(new_link[-2])[-1], \
new_link[-1])
new_col_name=labelname
img_data = {}
plot_label=labelname
if data.has_key('support_pcoas'):
matrix_average, matrix_low, matrix_high, eigval_average, m_names = \
summarize_pcoas(data['coord'], data['support_pcoas'],
method=data['ellipsoid_method'])
data['coord'] = \
(m_names,matrix_average,data['coord'][2],data['coord'][3])
for i in range(len(m_names)):
sample_location[m_names[i]]=i
else:
matrix_average = None
matrix_low = None
matrix_high = None
eigval_average = None
m_names = None
iterator=0
for coord_tup in coord_tups: # change, if you want more thatn one PCoA plot! (i.e involving PC3)
if isarray(matrix_low) and isarray(matrix_high) and \
isarray(matrix_average):
coord_1r=asarray(matrix_low)
coord_2r=asarray(matrix_high)
mat_ave=asarray(matrix_average)
else:
coord_1r=None
coord_2r=None
mat_ave=None
sample_location=None
coord_1, coord_2 = coord_tup
img_data[coord_tup] = draw_pcoa_graph(plot_label,data_file_dir_path,
data_file_link,coord_1,coord_2,
coord_1r, coord_2r, mat_ave,\
sample_location,
data,prefs,groups,colors,
background_color,label_color,
data_colors,data_color_order,
generate_eps=True)
radiobuttons.append(RADIO % (data_file_link, labelname))
if i == 0: ## only create first table!
out_table += TABLE_HTML % (labelname,
"<br>".join(img_data[("1", "2")]),
"<br>".join(img_data[("3", "2")]),
"<br>".join(img_data[("1", "3")]))
if generate_scree:
data_file_dir_path = get_random_directory_name(output_dir = data_dir_path)
new_link = os.path.split(data_file_dir_path)
data_file_link = os.path.join('.', os.path.split(new_link[-2])[-1], new_link[-1])
img_src, download_link = draw_scree_graph(data_file_dir_path, data_file_link, background_color,
label_color, generate_eps = True, data = data)
out_table += SCREE_TABLE_HTML % ("<br>".join((img_src, download_link)))
out_table = "\n".join(radiobuttons) + out_table
outfile = create_html_filename(filename,'.html')
outfile = os.path.join(html_dir_path,outfile)
write_html_file(out_table,outfile)
def preprocess_coords_file(coords_header,
coords_data,
coords_eigenvals,
coords_pct,
mapping_header,
mapping_data,
custom_axes=None,
jackknifing_method=None,
is_comparison=False):
"""Process a PCoA data and handle customizations in the contents
Inputs:
coords_header: list of sample identifiers in the PCoA file _or_ list of
lists with sample identifiers for each coordinate file (if jackknifing or
comparing plots)
coords_data: matrix of coordinates in the PCoA file _or_ list of numpy
arrays with coordinates for each file (if jackknifing or comparing plots)
coords_eigenvals: numpy array with eigenvalues for the coordinates file _or_
list of numpy arrays with the eigenvalues (if jackknifing or comparing plots
)
coords_pct: numpy array with a the percent explained by each principal
coordinates axis _or_ a list of lists with numpy arrays (if jackknifing or
comparing plots)
mapping_header: mapping file headers names
mapping_data: mapping file data
custom_axes: name of the mapping data fields to add to coords_data
jackknifing_method: one of 'sdev' or 'IRQ', defaults to None, for more info
see qiime.util.summarize_pcoas
is_comparison: whether or not the inputs should be considered as the ones
for a comparison plot
Outputs:
coords_header: list of sample identifiers in the PCoA file
coords_data: matrix of coordinates in the PCoA file with custom_axes if
provided
coords_eigenvalues: either the eigenvalues of the input coordinates or the
average eigenvalues of the multiple coords that were passed in
coords_pct: list of percents explained by each axis as given by the master
coordinates i. e. the center around where the values revolve
coords_low: coordinates representing the lower edges of an ellipse; None if
no jackknifing is applied
coords_high: coordinates representing the highere edges of an ellipse; None
if no jackknifing is applied
clones: total number of input files
This controller function handles any customization that has to be done to
the PCoA data prior to the formatting. Note that the first element in each
list (coords, headers, eigenvalues & percents) will be considered the master
set of coordinates.
Raises: AssertionError if a comparison plot is requested but a list of data
is not passed as input
"""
# prevent obscure and obfuscated errors
if is_comparison:
assert type(coords_data) == list, "Cannot process a comparison with "+\
"the data from a single coordinates file"
mapping_file = [mapping_header] + mapping_data
coords_file = [coords_header, coords_data]
# number PCoA files; zero for any case except for comparison plots
clones = 0
if custom_axes and type(coords_data) == ndarray:
# sequence ported from qiime/scripts/make_3d_plots.py @ 9115351
get_custom_coords(custom_axes, mapping_file, coords_file)
remove_nans(coords_file)
scale_custom_coords(custom_axes, coords_file)
elif type(coords_data) == list and is_comparison == False:
# take the first pcoa file as the master set of coordinates
master_pcoa = [
coords_header[0], coords_data[0], coords_eigenvals[0],
coords_pct[0]
]
# support pcoas must be a list of lists where each list contain
# all the elements that compose a coordinates file
support_pcoas = [[h, d, e, p] for h, d, e, p in zip(
coords_header, coords_data, coords_eigenvals, coords_pct)]
# do not apply procrustes, at least not for now
coords_data, coords_low, coords_high, eigenvalues_average,\
identifiers = summarize_pcoas(master_pcoa, support_pcoas,
method=jackknifing_method, apply_procrustes=False)
# custom axes and jackknifing is a tricky thing to do, you only have to
# add the custom values to the master file which is represented as the
# coords_data return value. Since there is really no variation in that
# axis then you have to change the values of coords_high and of
# coords_low to something really small so that WebGL work properly
if custom_axes:
coords_file = [master_pcoa[0], coords_data]
get_custom_coords(custom_axes, mapping_file, coords_file)
remove_nans(coords_file)
scale_custom_coords(custom_axes, coords_file)
# this opens support for as many custom axes as needed
axes = len(custom_axes)
coords_low[:, 0:axes] = zeros([coords_low.shape[0], axes])
coords_high[:,
0:axes] = ones([coords_high.shape[0], axes]) * 0.00001
coords_data = coords_file[1]
# return a value containing coords_low and coords_high
return identifiers, coords_data, eigenvalues_average, master_pcoa[3],\
coords_low, coords_high, clones
# comparison plots are processed almost individually
elif type(coords_data) == list and is_comparison:
# indicates the number of files that were totally processed so other
# functions/APIs are aware of how many times to replicate the metadata
clones = len(coords_data)
out_headers, out_coords = [], []
for index in range(0, clones):
headers_i = coords_header[index]
coords_i = coords_data[index]
# tag each header with the the number in which those coords came in
out_headers.extend(
[element + '_%d' % index for element in headers_i])
if index == 0:
# numpy can only stack things if they have the same shape
out_coords = coords_i
# the eigenvalues and percents explained are really the ones
# belonging to the the first set of coordinates that was passed
coords_eigenvals = coords_eigenvals[index]
coords_pct = coords_pct[index]
else:
out_coords = vstack((out_coords, coords_i))
coords_file = [out_headers, out_coords]
if custom_axes:
# this condition deals with the fact that in order for the custom
# axes to be added into the original coordinates, we have to add the
# suffix for the sample identifiers that the coordinates have
if clones:
out_data = []
for index in range(0, clones):
out_data.extend([[element[0] + '_%d' % index] +
element[1::] for element in mapping_data])
mapping_file = [mapping_header] + out_data
# sequence ported from qiime/scripts/make_3d_plots.py @ 9115351
get_custom_coords(custom_axes, mapping_file, coords_file)
remove_nans(coords_file)
scale_custom_coords(custom_axes, coords_file)
# if no coords summary is applied, return None in the corresponding values
# note that the value of clones will be != 0 for a comparison plot
return coords_file[0], coords_file[1], coords_eigenvals, coords_pct, None,\
None, clones
def preprocess_coords_file(
coords_header,
coords_data,
coords_eigenvals,
coords_pct,
mapping_header,
mapping_data,
custom_axes=None,
jackknifing_method=None,
is_comparison=False,
):
"""Process a PCoA data and handle customizations in the contents
Inputs:
coords_header: list of sample identifiers in the PCoA file _or_ list of
lists with sample identifiers for each coordinate file (if jackknifing or
comparing plots)
coords_data: matrix of coordinates in the PCoA file _or_ list of numpy
arrays with coordinates for each file (if jackknifing or comparing plots)
coords_eigenvals: numpy array with eigenvalues for the coordinates file _or_
list of numpy arrays with the eigenvalues (if jackknifing or comparing plots
)
coords_pct: numpy array with a the percent explained by each principal
coordinates axis _or_ a list of lists with numpy arrays (if jackknifing or
comparing plots)
mapping_header: mapping file headers names
mapping_data: mapping file data
custom_axes: name of the mapping data fields to add to coords_data
jackknifing_method: one of 'sdev' or 'IRQ', defaults to None, for more info
see qiime.util.summarize_pcoas
is_comparison: whether or not the inputs should be considered as the ones
for a comparison plot
Outputs:
coords_header: list of sample identifiers in the PCoA file
coords_data: matrix of coordinates in the PCoA file with custom_axes if
provided
coords_eigenvalues: either the eigenvalues of the input coordinates or the
average eigenvalues of the multiple coords that were passed in
coords_pct: list of percents explained by each axis as given by the master
coordinates i. e. the center around where the values revolve
coords_low: coordinates representing the lower edges of an ellipse; None if
no jackknifing is applied
coords_high: coordinates representing the highere edges of an ellipse; None
if no jackknifing is applied
clones: total number of input files
This controller function handles any customization that has to be done to
the PCoA data prior to the formatting. Note that the first element in each
list (coords, headers, eigenvalues & percents) will be considered the master
set of coordinates.
Raises: AssertionError if a comparison plot is requested but a list of data
is not passed as input
"""
# prevent obscure and obfuscated errors
if is_comparison:
assert type(coords_data) == list, (
"Cannot process a comparison with " + "the data from a single coordinates file"
)
mapping_file = [mapping_header] + mapping_data
coords_file = [coords_header, coords_data]
# number PCoA files; zero for any case except for comparison plots
clones = 0
if custom_axes and type(coords_data) == ndarray:
# sequence ported from qiime/scripts/make_3d_plots.py @ 9115351
get_custom_coords(custom_axes, mapping_file, coords_file)
remove_nans(coords_file)
scale_custom_coords(custom_axes, coords_file)
elif type(coords_data) == list and is_comparison == False:
# take the first pcoa file as the master set of coordinates
master_pcoa = [coords_header[0], coords_data[0], coords_eigenvals[0], coords_pct[0]]
# support pcoas must be a list of lists where each list contain
# all the elements that compose a coordinates file
support_pcoas = [[h, d, e, p] for h, d, e, p in zip(coords_header, coords_data, coords_eigenvals, coords_pct)]
# do not apply procrustes, at least not for now
coords_data, coords_low, coords_high, eigenvalues_average, identifiers = summarize_pcoas(
master_pcoa, support_pcoas, method=jackknifing_method, apply_procrustes=False
)
# custom axes and jackknifing is a tricky thing to do, you only have to
# add the custom values to the master file which is represented as the
# coords_data return value. Since there is really no variation in that
# axis then you have to change the values of coords_high and of
# coords_low to something really small so that WebGL work properly
if custom_axes:
coords_file = [master_pcoa[0], coords_data]
get_custom_coords(custom_axes, mapping_file, coords_file)
remove_nans(coords_file)
scale_custom_coords(custom_axes, coords_file)
# this opens support for as many custom axes as needed
axes = len(custom_axes)
coords_low[:, 0:axes] = zeros([coords_low.shape[0], axes])
coords_high[:, 0:axes] = ones([coords_high.shape[0], axes]) * 0.00001
coords_data = coords_file[1]
# return a value containing coords_low and coords_high
return identifiers, coords_data, eigenvalues_average, master_pcoa[3], coords_low, coords_high, clones
# comparison plots are processed almost individually
elif type(coords_data) == list and is_comparison:
# indicates the number of files that were totally processed so other
# functions/APIs are aware of how many times to replicate the metadata
clones = len(coords_data)
out_headers, out_coords = [], []
for index in range(0, clones):
headers_i = coords_header[index]
coords_i = coords_data[index]
# tag each header with the the number in which those coords came in
out_headers.extend([element + "_%d" % index for element in headers_i])
if index == 0:
# numpy can only stack things if they have the same shape
out_coords = coords_i
# the eigenvalues and percents explained are really the ones
# belonging to the the first set of coordinates that was passed
coords_eigenvals = coords_eigenvals[index]
coords_pct = coords_pct[index]
else:
out_coords = vstack((out_coords, coords_i))
coords_file = [out_headers, out_coords]
if custom_axes:
# this condition deals with the fact that in order for the custom
# axes to be added into the original coordinates, we have to add the
# suffix for the sample identifiers that the coordinates have
if clones:
out_data = []
for index in range(0, clones):
out_data.extend([[element[0] + "_%d" % index] + element[1::] for element in mapping_data])
mapping_file = [mapping_header] + out_data
# sequence ported from qiime/scripts/make_3d_plots.py @ 9115351
get_custom_coords(custom_axes, mapping_file, coords_file)
remove_nans(coords_file)
scale_custom_coords(custom_axes, coords_file)
# if no coords summary is applied, return None in the corresponding values
# note that the value of clones will be != 0 for a comparison plot
return coords_file[0], coords_file[1], coords_eigenvals, coords_pct, None, None, clones