def draw_scatterplot(props,xy_coords,x_len,y_len,size,background_color,
label_color,sample_location,alpha):
"""Create scatterplot figure"""
fig = figure(figsize=(x_len,y_len))
xPC=int(props['xlabel'][1:2])
yPC=int(props['ylabel'][1:2])
sorted_keys = xy_coords.keys()
scatters = {}
size_ct = shape_ct = 0
xPC=xPC-1
yPC=yPC-1
#Iterate through coords and add points to the scatterplot
for s_label in sorted_keys:
s_data = xy_coords[s_label]
if s_data[0]==[]:
pass
else:
c = s_data[3]
m = s_data[4][0]
ax = fig.add_subplot(111,axisbg=background_color)
#set tick colors and width
for line in ax.yaxis.get_ticklines():
# line is a matplotlib.lines.Line2D instance
line.set_color(label_color)
line.set_markeredgewidth(1)
for line in ax.xaxis.get_ticklines():
# line is a matplotlib.lines.Line2D instance
line.set_color(label_color)
line.set_markeredgewidth(1)
if isarray(s_data[5][0]) and isarray(s_data[6][0]) and isarray(s_data[7][0]):
matrix_low=s_data[5][0]
matrix_high=s_data[6][0]
ellipse_ave=s_data[7][0]
ellipse_x=[ellipse_ave[sample_location[s_label], xPC]]
ellipse_y=[ellipse_ave[sample_location[s_label], yPC]]
width=[fabs(matrix_high[sample_location[s_label], xPC] - \
matrix_low[sample_location[s_label], xPC])]
height=[fabs(matrix_high[sample_location[s_label], yPC] - \
matrix_low[sample_location[s_label], yPC])]
sc_plot = scatter_ellipse(ax,ellipse_x, \
ellipse_y, width, height, c=c, a=0.0, \
alpha=alpha)
sc_plot.scatter(ellipse_x, ellipse_y, c=c, marker=m, \
alpha=1.0)
else:
sc_plot = ax.scatter(s_data[0], s_data[1], c=c, marker=m, \
alpha=1.0,s=size, linewidth=1,edgecolor=c)
size_ct += 1
shape_ct += 1
scatters[s_label] = sc_plot
return sc_plot
def test_isarray(self):
"isarray: tests the isarray function"
test1 = asarray('Test')
test2 = 'Test'
exp1 = True
exp2 = False
self.assertEqual(isarray(test1), exp1)
self.assertEqual(isarray(test2), exp2)
def test_isarray(self):
"isarray: tests the isarray function"
test1=asarray('Test')
test2 = 'Test'
exp1 = True
exp2 = False
self.assertEqual(isarray(test1),exp1)
self.assertEqual(isarray(test2),exp2)
def draw_scatterplot(props, xy_coords, x_len, y_len, size, background_color, label_color, sample_location, alpha):
"""Create scatterplot figure"""
fig = figure(figsize=(x_len, y_len))
xPC = int(props["xlabel"][2:3])
yPC = int(props["ylabel"][2:3])
sorted_keys = xy_coords.keys()
scatters = {}
size_ct = shape_ct = 0
xPC = xPC - 1
yPC = yPC - 1
# Iterate through coords and add points to the scatterplot
for s_label in sorted_keys:
s_data = xy_coords[s_label]
if s_data[0] == []:
pass
else:
c = s_data[3]
m = s_data[4][0]
ax = fig.add_subplot(111, axisbg=background_color)
# set tick colors and width
for line in ax.yaxis.get_ticklines():
# line is a matplotlib.lines.Line2D instance
line.set_color(label_color)
line.set_markeredgewidth(1)
for line in ax.xaxis.get_ticklines():
# line is a matplotlib.lines.Line2D instance
line.set_color(label_color)
line.set_markeredgewidth(1)
if isarray(s_data[5][0]) and isarray(s_data[6][0]) and isarray(s_data[7][0]):
matrix_low = s_data[5][0]
matrix_high = s_data[6][0]
ellipse_ave = s_data[7][0]
ellipse_x = [ellipse_ave[sample_location[s_label], xPC]]
ellipse_y = [ellipse_ave[sample_location[s_label], yPC]]
width = [fabs(matrix_high[sample_location[s_label], xPC] - matrix_low[sample_location[s_label], xPC])]
height = [fabs(matrix_high[sample_location[s_label], yPC] - matrix_low[sample_location[s_label], yPC])]
sc_plot = scatter_ellipse(ax, ellipse_x, ellipse_y, width, height, c=c, a=0.0, alpha=alpha)
sc_plot.scatter(ellipse_x, ellipse_y, c=c, marker=m, alpha=1.0)
else:
sc_plot = ax.scatter(s_data[0], s_data[1], c=c, marker=m, alpha=1.0, s=size, linewidth=1, edgecolor=c)
size_ct += 1
shape_ct += 1
scatters[s_label] = sc_plot
return sc_plot
def extract_and_color_xy_coords(p1d,p2d,p1dr,p2dr,mat_ave,colors, data_colors, \
groups, coords, tooltipDict):
"""Extract coords from appropriate columns and attach their \
corresponding colors based on the group"""
xy_coords = {}
shape_ct = 0
for group_name, ids in (groups.items()):
x=0
if isinstance(colors[group_name], list):
color = colors[group_name]#[c if c.startswith("#") else data_colors[c].toHex() for c in colors[group_name]]
else:
color = colors[group_name] if colors[group_name].startswith("#") else data_colors[colors[group_name]].toHex() ## so I can use my own hex colors, defined in prefs.txt
#color = data_colors[colors[group_name]].toHex()
m =shape[shape_ct % len(shape)]
shape_ct += 1
for id_ in (ids):
cur_labs = []
cur_x = []
cur_y = []
cur_color = []
cur_shape=[]
cur_1r=[]
cur_2r=[]
new_mat_ave=[]
if id_ in coords['pc vector number']:
cur_labs.append(tooltipDict[id_]) #id_+': '+group_name)
cur_x.append(p2d[id_])
cur_y.append(p1d[id_])
#pdb.set_trace()
cur_color.append(color)
cur_shape.append(m)
if isarray(p2dr) and isarray(p1dr) and isarray(mat_ave):
cur_1r.append(p1dr)
cur_2r.append(p2dr)
new_mat_ave.append(mat_ave)
else:
cur_1r=[None]
cur_2r=[None]
new_mat_ave=[None]
xy_coords["%s" % id_] = (cur_x, cur_y, cur_labs, cur_color, \
cur_shape, cur_1r, cur_2r,new_mat_ave)
return xy_coords
def extract_and_color_xy_coords(p1d,p2d,p1dr,p2dr,mat_ave,colors, data_colors, \
groups, coords, tooltipDict):
"""Extract coords from appropriate columns and attach their \
corresponding colors based on the group"""
xy_coords = {}
shape_ct = 0
for group_name, ids in (groups.items()):
x=0
if isinstance(colors[group_name], list):
color = colors[group_name]#[c if c.startswith("#") else data_colors[c].toHex() for c in colors[group_name]]
else:
color = colors[group_name] if colors[group_name].startswith("#") else data_colors[colors[group_name]].toHex() ## so I can use my own hex colors, defined in prefs.txt
#color = data_colors[colors[group_name]].toHex()
m =shape[shape_ct % len(shape)]
shape_ct += 1
for id_ in (ids):
cur_labs = []
cur_x = []
cur_y = []
cur_color = []
cur_shape=[]
cur_1r=[]
cur_2r=[]
new_mat_ave=[]
if id_ in coords['pc vector number']:
cur_labs.append(tooltipDict[id_]) #id_+': '+group_name)
cur_x.append(p2d[id_])
cur_y.append(p1d[id_])
#pdb.set_trace()
cur_color.append(color)
cur_shape.append(m)
if isarray(p2dr) and isarray(p1dr) and isarray(mat_ave):
cur_1r.append(p1dr)
cur_2r.append(p2dr)
new_mat_ave.append(mat_ave)
else:
cur_1r=[None]
cur_2r=[None]
new_mat_ave=[None]
xy_coords["%s" % id_] = (cur_x, cur_y, cur_labs, cur_color, \
cur_shape, cur_1r, cur_2r,new_mat_ave)
return xy_coords
def extract_and_color_xy_coords(
p1d, p2d, p1dr, p2dr, mat_ave, colors, data_colors,
groups, coords):
"""Extract coords from appropriate columns and attach their \
corresponding colors based on the group"""
xy_coords = {}
shape_ct = 0
for group_name, ids in (groups.items()):
x = 0
color = data_colors[colors[group_name]].toHex()
m = shape[shape_ct % len(shape)]
shape_ct += 1
for id_ in (ids):
cur_labs = []
cur_x = []
cur_y = []
cur_color = []
cur_shape = []
cur_1r = []
cur_2r = []
new_mat_ave = []
if id_ in coords['pc vector number']:
cur_labs.append(id_ + ': ' + group_name)
cur_x.append(p2d[id_])
cur_y.append(p1d[id_])
cur_color.append(color)
cur_shape.append(m)
if isarray(p2dr) and isarray(p1dr) and isarray(mat_ave):
cur_1r.append(p1dr)
cur_2r.append(p2dr)
new_mat_ave.append(mat_ave)
else:
cur_1r = [None]
cur_2r = [None]
new_mat_ave = [None]
xy_coords["%s" % id_] = (cur_x, cur_y, cur_labs, cur_color,
cur_shape, cur_1r, cur_2r, new_mat_ave)
return xy_coords
def extract_and_color_xy_coords(
p1d, p2d, p1dr, p2dr, mat_ave, colors, data_colors,
groups, coords):
"""Extract coords from appropriate columns and attach their \
corresponding colors based on the group"""
xy_coords = {}
shape_ct = 0
for group_name, ids in (groups.items()):
x = 0
color = data_colors[colors[group_name]].toHex()
m = shape[shape_ct % len(shape)]
shape_ct += 1
for id_ in (ids):
cur_labs = []
cur_x = []
cur_y = []
cur_color = []
cur_shape = []
cur_1r = []
cur_2r = []
new_mat_ave = []
if id_ in coords['pc vector number']:
cur_labs.append(id_ + ': ' + group_name)
cur_x.append(p2d[id_])
cur_y.append(p1d[id_])
cur_color.append(color)
cur_shape.append(m)
if isarray(p2dr) and isarray(p1dr) and isarray(mat_ave):
cur_1r.append(p1dr)
cur_2r.append(p2dr)
new_mat_ave.append(mat_ave)
else:
cur_1r = [None]
cur_2r = [None]
new_mat_ave = [None]
xy_coords["%s" % id_] = (cur_x, cur_y, cur_labs, cur_color,
cur_shape, cur_1r, cur_2r, new_mat_ave)
return xy_coords
def draw_pcoa_graph(plot_label, 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):
"""Draw PCoA graphs"""
coords,pct_var=convert_coord_data_to_dict(data)
mapping = data['map']
tooltipDict = dict([(sample[0], "%s/%s/%s<BR>%s<BR>%s<BR>%s"%(sample[0],sample[1],sample[3], sample[2], sample[4], sample[5][:120])) for sample in mapping])
if coord_1 not in coords:
raise ValueError, "Principal coordinate: %s not available." % coord_1
if coord_2 not in coords:
raise ValueError, "Principal coordinate: %s not available." % coord_2
#Handle matplotlib scale bug when all coords are 0.0
if not len([x for x in map(float, coords[coord_2]) if x != 0.0]):
for ix in range(len(coords[coord_2])):
coords[coord_2][ix] = '1e-255'
if not len([x for x in map(float, coords[coord_1]) if x != 0.0]):
for ix in range(len(coords[coord_1])):
coords[coord_1][ix] = '1e-255'
#Write figure labels
props = {}
props["title"] = "PCoA - PC%s vs PC%s" % (coord_1, coord_2)
props["imgID"] = "PC%s%s" % (coord_1, coord_2)
props["ylabel"] = "PC%s - Percent variation explained %.2f%%" \
% (coord_2, float(pct_var[coord_2]))
props["xlabel"] = "PC%s - Percent variation explained %.2f%%" \
% (coord_1, float(pct_var[coord_1]))
labels = coords['pc vector number']
p1 = map(float, coords[coord_2])
p2 = map(float, coords[coord_1])
if isarray(coord_1r) and isarray(coord_2r) and isarray(mat_ave):
p1r = coord_2r
p2r = coord_1r
else:
p1r = None
p2r = None
mat_ave=None
if len(p1) != len(p2):
raise ValueError, "Principal coordinate vectors unequal length."
p1d = dict(zip(labels, p1))
p2d = dict(zip(labels, p2))
alpha=data['alpha']
xy_coords = extract_and_color_xy_coords(p1d, p2d, p1r, p2r,mat_ave,colors, \
data_colors, groups, coords, tooltipDict)
img_src, img_map, eps_link = make_interactive_scatter(plot_label,dir_path,
data_file_link,background_color,label_color,
sample_location,alpha,
xy_coords=xy_coords,props=props,x_len=12,
y_len=12,size=20,draw_axes=True,
generate_eps=generate_eps)
return img_src + img_map, eps_link
def draw_scatterplot(props,xy_coords,x_len,y_len,size,background_color,
label_color,sample_location,alpha):
"""Create scatterplot figure"""
print "Figsize", x_len, y_len
fig = figure(figsize=(x_len,y_len))
xPC=int(props['xlabel'][2:3])
yPC=int(props['ylabel'][2:3])
sorted_keys = xy_coords.keys()
scatters = {}
size_ct = shape_ct = 0
xPC=xPC-1
yPC=yPC-1
#Iterate through coords and add points to the scatterplot
for s_label in sorted_keys:
s_data = xy_coords[s_label]
if s_data[0]==[]:
pass
else:
c = s_data[3]
m = s_data[4][0]
ax = fig.add_subplot(111,axisbg=background_color)
#set tick colors and width
for line in ax.yaxis.get_ticklines():
# line is a matplotlib.lines.Line2D instance
line.set_color(label_color)
line.set_markeredgewidth(1)
for line in ax.xaxis.get_ticklines():
# line is a matplotlib.lines.Line2D instance
line.set_color(label_color)
line.set_markeredgewidth(1)
if isarray(s_data[5][0]) and isarray(s_data[6][0]) and \
isarray(s_data[7][0]):
matrix_low=s_data[5][0]
matrix_high=s_data[6][0]
ellipse_ave=s_data[7][0]
ellipse_x=[ellipse_ave[sample_location[s_label], xPC]]
ellipse_y=[ellipse_ave[sample_location[s_label], yPC]]
width=[fabs(matrix_high[sample_location[s_label], xPC] - \
matrix_low[sample_location[s_label], xPC])]
height=[fabs(matrix_high[sample_location[s_label], yPC] - \
matrix_low[sample_location[s_label], yPC])]
sc_plot = scatter_ellipse(ax,ellipse_x, \
ellipse_y, width, height, c=c, a=0.0, \
alpha=alpha)
sc_plot.scatter(ellipse_x, ellipse_y, c=c, marker=m, \
alpha=1.0)
else:
#pdb.set_trace()
if s_data[2][0].split("/")[0] == "Sabkha":
ax.scatter(s_data[0], s_data[1], c='r', marker='o',
alpha=0.5,s=18, linewidth=2,edgecolor='k')
colorlist = flatten(c)[:3] if c and isinstance(c[0], list) else c
for idx, col in enumerate(colorlist):
dotsize = 15-6*idx
try:
sc_plot = ax.scatter(s_data[0], s_data[1], c=col, marker='o',
alpha=1,s=dotsize, linewidth=0) #,edgecolor=c)
except:
pdb.set_trace()
size_ct += 1
shape_ct += 1
scatters[s_label] = sc_plot
return sc_plot
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 draw_pcoa_graph(plot_label, 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):
"""Draw PCoA graphs"""
coords, pct_var = convert_coord_data_to_dict(data)
mapping = data['map']
if coord_1 not in coords:
raise ValueError, "Principal coordinate: %s not available." % coord_1
if coord_2 not in coords:
raise ValueError, "Principal coordinate: %s not available." % coord_2
#Handle matplotlib scale bug when all coords are 0.0
if not len([x for x in map(float, coords[coord_2]) if x != 0.0]):
for ix in range(len(coords[coord_2])):
coords[coord_2][ix] = '1e-255'
if not len([x for x in map(float, coords[coord_1]) if x != 0.0]):
for ix in range(len(coords[coord_1])):
coords[coord_1][ix] = '1e-255'
#Write figure labels
props = {}
props["title"] = "PCoA - PC%s vs PC%s" % (coord_1, coord_2)
props["ylabel"] = "PC%s - Percent variation explained %.2f%%" \
% (coord_2, float(pct_var[coord_2]))
props["xlabel"] = "PC%s - Percent variation explained %.2f%%" \
% (coord_1, float(pct_var[coord_1]))
labels = coords['pc vector number']
p1 = map(float, coords[coord_2])
p2 = map(float, coords[coord_1])
if isarray(coord_1r) and isarray(coord_2r) and isarray(mat_ave):
p1r = coord_2r
p2r = coord_1r
else:
p1r = None
p2r = None
mat_ave = None
if len(p1) != len(p2):
raise ValueError, "Principal coordinate vectors unequal length."
p1d = dict(zip(labels, p1))
p2d = dict(zip(labels, p2))
alpha = data['alpha']
xy_coords = extract_and_color_xy_coords(p1d, p2d, p1r, p2r,mat_ave,colors, \
data_colors, groups, coords)
img_src, img_map, eps_link = make_interactive_scatter(
plot_label,
dir_path,
data_file_link,
background_color,
label_color,
sample_location,
alpha,
xy_coords=xy_coords,
props=props,
x_len=4.5,
y_len=4.5,
size=20,
draw_axes=True,
generate_eps=generate_eps)
return img_src + img_map, eps_link
def draw_scatterplot(props,xy_coords,x_len,y_len,size,background_color,
label_color,sample_location,alpha):
"""Create scatterplot figure"""
print "Figsize", x_len, y_len
fig = figure(figsize=(x_len,y_len))
xPC=int(props['xlabel'][2:3])
yPC=int(props['ylabel'][2:3])
sorted_keys = xy_coords.keys()
scatters = {}
size_ct = shape_ct = 0
xPC=xPC-1
yPC=yPC-1
#Iterate through coords and add points to the scatterplot
for s_label in sorted_keys:
s_data = xy_coords[s_label]
if s_data[0]==[]:
pass
else:
c = s_data[3]
m = s_data[4][0]
ax = fig.add_subplot(111,axisbg=background_color)
#set tick colors and width
for line in ax.yaxis.get_ticklines():
# line is a matplotlib.lines.Line2D instance
line.set_color(label_color)
line.set_markeredgewidth(1)
for line in ax.xaxis.get_ticklines():
# line is a matplotlib.lines.Line2D instance
line.set_color(label_color)
line.set_markeredgewidth(1)
if isarray(s_data[5][0]) and isarray(s_data[6][0]) and \
isarray(s_data[7][0]):
matrix_low=s_data[5][0]
matrix_high=s_data[6][0]
ellipse_ave=s_data[7][0]
ellipse_x=[ellipse_ave[sample_location[s_label], xPC]]
ellipse_y=[ellipse_ave[sample_location[s_label], yPC]]
width=[fabs(matrix_high[sample_location[s_label], xPC] - \
matrix_low[sample_location[s_label], xPC])]
height=[fabs(matrix_high[sample_location[s_label], yPC] - \
matrix_low[sample_location[s_label], yPC])]
sc_plot = scatter_ellipse(ax,ellipse_x, \
ellipse_y, width, height, c=c, a=0.0, \
alpha=alpha)
sc_plot.scatter(ellipse_x, ellipse_y, c=c, marker=m, \
alpha=1.0)
else:
#pdb.set_trace()
if s_data[2][0].split("/")[0] == "Sabkha":
ax.scatter(s_data[0], s_data[1], c='r', marker='o',
alpha=0.5,s=18, linewidth=2,edgecolor='k')
colorlist = flatten(c)[:3] if c and isinstance(c[0], list) else c
for idx, col in enumerate(colorlist):
dotsize = 15-6*idx
try:
sc_plot = ax.scatter(s_data[0], s_data[1], c=col, marker='o',
alpha=1,s=dotsize, linewidth=0) #,edgecolor=c)
except:
pdb.set_trace()
size_ct += 1
shape_ct += 1
scatters[s_label] = sc_plot
return sc_plot
def draw_pcoa_graph(plot_label, 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,
pct_variation_below_one=False):
"""Draw PCoA graphs"""
coords, pct_var = convert_coord_data_to_dict(data)
if coord_1 not in coords:
raise ValueError("Principal coordinate: %s not available." % coord_1)
if coord_2 not in coords:
raise ValueError("Principal coordinate: %s not available." % coord_2)
# Handle matplotlib scale bug when all coords are 0.0
if not len([x for x in map(float, coords[coord_2]) if x != 0.0]):
for ix in range(len(coords[coord_2])):
coords[coord_2][ix] = '1e-255'
if not len([x for x in map(float, coords[coord_1]) if x != 0.0]):
for ix in range(len(coords[coord_1])):
coords[coord_1][ix] = '1e-255'
# Write figure labels
pct_exp1 = float(pct_var[coord_1])
pct_exp2 = float(pct_var[coord_2])
if float(pct_var['1']) < 1 and not pct_variation_below_one:
pct_exp1 *= 100
pct_exp2 *= 100
props = {}
props["title"] = "PCoA - PC%s vs PC%s" % (coord_1, coord_2)
props["ylabel"] = "PC%s - Percent variation explained %.2f%%" \
% (coord_2, pct_exp2)
props["xlabel"] = "PC%s - Percent variation explained %.2f%%" \
% (coord_1, pct_exp1)
labels = coords['pc vector number']
p1 = map(float, coords[coord_2])
p2 = map(float, coords[coord_1])
if isarray(coord_1r) and isarray(coord_2r) and isarray(mat_ave):
p1r = coord_2r
p2r = coord_1r
else:
p1r = None
p2r = None
mat_ave = None
if len(p1) != len(p2):
raise ValueError("Principal coordinate vectors unequal length.")
p1d = dict(zip(labels, p1))
p2d = dict(zip(labels, p2))
alpha = data['alpha']
xy_coords = extract_and_color_xy_coords(
p1d, p2d, p1r, p2r, mat_ave, colors,
data_colors, groups, coords)
img_src, img_map, eps_link = make_interactive_scatter(
plot_label, dir_path,
data_file_link, background_color, label_color,
sample_location, alpha,
xy_coords=xy_coords, props=props, x_len=4.5,
y_len=4.5, size=20, draw_axes=True,
generate_eps=generate_eps)
return img_src + img_map, eps_link
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)
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 = 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)