def peruse_forward_scores():
_, reco_info = seqfileopener.get_seqfile_info(simfname, is_data=False) #, n_max_queries=10000)
logprobs, partialcorr_logprobs, corr_logprobs = OrderedDict(), OrderedDict(), OrderedDict()
for n_set in n_set_list:
print n_set
# if n_set != 5:
# continue
logprobs[n_set], partialcorr_logprobs[n_set], corr_logprobs[n_set] = OrderedDict(), OrderedDict(), OrderedDict()
with open(outputdir + '/' + str(n_set) + '-forward.csv') as csvfile:
reader = csv.DictReader(csvfile)
for line in reader:
uidlist = line['unique_ids'].split(':')
assert utils.from_same_event(reco_info, uidlist)
reco_id = reco_info[uidlist[0]]['reco_id']
if reco_id in logprobs[n_set]:
raise Exception('already had %s' % reco_id)
logprobs[n_set][reco_id] = float(line['logprob'])
factor = 1. / n_set
partialcorr_logprobs[n_set][reco_id] = factor * float(line['logprob'])
factor = (1. - 0.24 / pow(float(n_set), 0.9)) / n_set
# factor = 1. / (0.77547824*n_set + 0.20327936)
corr_logprobs[n_set][reco_id] = factor * float(line['logprob'])
i_baseline = -1
deviations = get_deviations(logprobs, i_baseline)
# fit_stuff(n_set_list, deviations)
partialcorr_deviations = get_deviations(partialcorr_logprobs, i_baseline)
signed_partialcorr_deviations = get_deviations(partialcorr_logprobs, i_baseline, signed=True)
corr_deviations = get_deviations(corr_logprobs, i_baseline)
signed_corr_deviations = get_deviations(corr_logprobs, i_baseline, signed=True)
import plotting
fig, ax = plotting.mpl_init()
ax.plot(n_set_list, deviations, marker='.')
plotting.mpl_finish(ax, baseplotdir, 'forwards', xlabel='N simultaneous seqs', ylabel='log prob deviation to ' + str(n_set_list[i_baseline])) #, ybounds=(-0.02, 0.02))
# fig, ax = plotting.mpl_init()
# ax.plot(n_set_list, partialcorr_deviations, marker='.')
# ax.plot([n_set_list[0], n_set_list[-1]], [0, 0])
# plotting.mpl_finish(ax, baseplotdir, 'partially-corrected-forwards', xlabel='N simultaneous seqs', ylabel='log prob deviation to ' + str(n_set_list[i_baseline])) #, ybounds=(-0.02, 0.02))
fig, ax = plotting.mpl_init()
ax.plot(n_set_list, partialcorr_deviations, marker='.', label='1/n (abs)')
ax.plot(n_set_list, signed_partialcorr_deviations, marker='.', label='1/n')
ax.plot(n_set_list, corr_deviations, marker='.', label='1/crap (abs)')
ax.plot(n_set_list, signed_corr_deviations, marker='.', label='1/crap')
ax.plot([n_set_list[0], n_set_list[-1]], [0, 0])
plotting.mpl_finish(ax, baseplotdir, 'corrected-forwards', xlabel='N simultaneous seqs', ylabel='log prob deviation to ' + str(n_set_list[i_baseline])) #, ybounds=(-0.02, 0.02))
fig, ax = plotting.mpl_init()
ax.plot(n_set_list, signed_corr_deviations, marker='.')
ax.plot([n_set_list[0], n_set_list[-1]], [0, 0])
plotting.mpl_finish(ax, baseplotdir, 'signed-corrected-forwards', xlabel='N simultaneous seqs', ylabel='log prob deviation to ' + str(n_set_list[i_baseline])) #, ybounds=(-0.02, 0.02))
def make_transition_plot(self, gene_name, model):
""" NOTE shares a lot with make_mutefreq_plot() in python/paramutils.py """
fig, ax = plotting.mpl_init()
fig.set_size_inches(plotting.plot_ratios[utils.get_region(gene_name)])
ibin = 0
print utils.color_gene(utils.unsanitize_name(gene_name))
legend_colors = set() # add a color to this the first time you plot it
for state in model.states:
# bin label
ax.text(-0.5 + ibin, -0.075, paramutils.simplify_state_name(state.name), rotation='vertical', size=8)
sorted_to_states = {}
for name in state.transitions.keys():
if name.find('IG') == 0 or name.find('TR') == 0:
sorted_to_states[name] = int(paramutils.simplify_state_name(name))
else:
sorted_to_states[name] = name
sorted_to_states = sorted(sorted_to_states.items(), key=operator.itemgetter(1))
total = 0.0
for to_state, simple_to_state in sorted_to_states:
prob = state.transitions[to_state]
alpha = 0.6
width = 3
if 'insert' in str(simple_to_state):
label = 'insert'
color = '#3498db' # blue
elif str(simple_to_state) == 'end':
label = 'end'
color = 'red'
else: # regional/internal states
assert to_state.find('IG') == 0 or to_state.find('TR') == 0
label = 'internal'
color = 'green'
label_to_use = None
if color not in legend_colors:
label_to_use = label
legend_colors.add(color)
# horizontal line at height total+prob
ax.plot([-0.5 + ibin, 0.5 + ibin], [total + prob, total + prob], color=color, linewidth=width, alpha=alpha, label=label_to_use)
# vertical line from total to total + prob
ax.plot([ibin, ibin], [total + 0.01, total + prob], color=color, alpha=alpha, linewidth=width)
midpoint = 0.5*(prob + 2*total)
# ax.text(ibin, midpoint, paramutils.simplify_state_name(to_state)) # nicely labels the midpoint of the chunk between lines, but there isn't really room for it
total += prob
ibin += 1
ax.get_xaxis().set_visible(False)
plotting.mpl_finish(ax, self.base_plotdir + '/transitions', gene_name, ybounds=(-0.01, 1.01), xbounds=(-3, len(model.states) + 3), leg_loc=(0.95, 0.1), adjust={'left' : 0.1, 'right' : 0.8}, leg_prop={'size' : 8})
def sample(self, n_vals, include_overflows=False, debug_plot=False): # draw <n_vals> random numbers from the x axis, according to the probabilities given by the bin contents NOTE similarity to recombinator.choose_vdj_combo()
assert not include_overflows # probably doesn't really make sense (since contents of overflows could've been from anywhere below/above, but we'd only return bin center), this is just a way to remind that it doesn't make sense
self.normalize(include_overflows=include_overflows) # if this is going to get called a lot with n_vals of 1, this would be slow, but otoh we *really* want to make sure things are normalized with include_overflows the same as it is here
centers = self.get_bin_centers()
pvals = numpy.random.uniform(0, 1, size=n_vals)
return_vals = [None for _ in pvals]
sum_prob, last_sum_prob = 0., 0.
for ibin in self.ibiniter(include_overflows):
sum_prob += self.bin_contents[ibin]
for iprob, pval in enumerate(pvals):
if pval < sum_prob and pval >= last_sum_prob:
return_vals[iprob] = centers[ibin]
last_sum_prob = sum_prob
assert return_vals.count(None) == 0
if debug_plot:
import plotting
fig, ax = plotting.mpl_init()
self.mpl_plot(ax, label='original')
shist = Hist(value_list=return_vals, init_int_bins=True)
shist.normalize(include_overflows=False)
shist.mpl_plot(ax, label='sampled', color='red')
plotting.mpl_finish(ax, '', 'tmp')
return return_vals
def make_transition_plot(self, gene_name, model):
""" NOTE shares a lot with make_mutefreq_plot() in python/paramutils.py """
fig, ax = plotting.mpl_init()
fig.set_size_inches(plotting.plot_ratios[utils.get_region(gene_name)])
ibin = 0
print utils.color_gene(utils.unsanitize_name(gene_name))
legend_colors = set() # add a color to this the first time you plot it
for state in model.states:
# bin label
ax.text(-0.5 + ibin, -0.075, paramutils.simplify_state_name(state.name), rotation='vertical', size=8)
sorted_to_states = {}
for name in state.transitions.keys():
if name.find('IG') == 0:
sorted_to_states[name] = int(paramutils.simplify_state_name(name))
else:
sorted_to_states[name] = name
sorted_to_states = sorted(sorted_to_states.items(), key=operator.itemgetter(1))
total = 0.0
for to_state, simple_to_state in sorted_to_states:
prob = state.transitions[to_state]
alpha = 0.6
width = 3
if 'insert' in str(simple_to_state):
label = 'insert'
color = '#3498db' # blue
elif str(simple_to_state) == 'end':
label = 'end'
color = 'red'
else: # regional/internal states
assert to_state.find('IG') == 0
label = 'internal'
color = 'green'
label_to_use = None
if color not in legend_colors:
label_to_use = label
legend_colors.add(color)
# horizontal line at height total+prob
ax.plot([-0.5 + ibin, 0.5 + ibin], [total + prob, total + prob], color=color, linewidth=width, alpha=alpha, label=label_to_use)
# vertical line from total to total + prob
ax.plot([ibin, ibin], [total + 0.01, total + prob], color=color, alpha=alpha, linewidth=width)
midpoint = 0.5*(prob + 2*total)
# ax.text(ibin, midpoint, paramutils.simplify_state_name(to_state)) # nicely labels the midpoint of the chunk between lines, but there isn't really room for it
total += prob
ibin += 1
ax.get_xaxis().set_visible(False)
plotting.mpl_finish(ax, self.base_plotdir + '/transitions', gene_name, ybounds=(-0.01, 1.01), xbounds=(-3, len(model.states) + 3), leg_loc=(0.95, 0.1), adjust={'left' : 0.1, 'right' : 0.8}, leg_prop={'size' : 8})
def peruse_naive_seqs():
from hist import Hist
# hall = Hist(n_set_list[-1], n_set_list[0] - 0.5, n_set_list[-1] + 0.5)
means = []
for n_set in n_set_list:
plotdir = baseplotdir + '/' + str(n_set)
hist = Hist(fname=plotdir + '/hmm/hamming_to_true_naive.csv')
print '%2d %.2f' % (n_set, hist.get_mean()),
# hall.set_ibin(hall.find_bin(n_set), hist.get_mean())
means.append(hist.get_mean())
import plotting
fig, ax = plotting.mpl_init()
# hall.mpl_plot(ax)
ax.plot(n_set_list, means, marker='.')
plotting.mpl_finish(ax, baseplotdir, 'means', xlabel='N simultaneous seqs', ylabel='mean hamming to true naive', ybounds=(0, None))
def plotheatmap(plotdir, plotname, difftype, genelist=None, genesets=None, title='', xtitle=''):
assert genelist is None or genesets is None
if genelist is not None:
smatrix = get_gene_pair_matrix(genelist, difftype)
xticklabels = [utils.summarize_gene_name(g) for g in genelist]
elif genesets is not None:
smatrix = get_gene_set_mean_matrix(genesets, difftype)
xticklabels = [sn for sn in genesets.keys()]
else:
raise Exception('no gene list specified')
assert len(smatrix) == len(smatrix[0]) # uh, I think I need this to be true
fig, ax = plotting.mpl_init()
plt.tick_params(axis='both', which='major', labelsize=7)
plt.gcf().subplots_adjust(bottom=0.14, left=0.18, right=0.95, top=0.92)
ax.xaxis.set_ticks_position('bottom')
ax.yaxis.set_ticks_position('left')
data = numpy.array(smatrix)
cmap = plt.cm.Blues #cm.get_cmap('jet')
cmap.set_under('w')
heatmap = ax.pcolor(data, cmap=cmap, vmin=0., vmax=0.5) #vmax=numpy.amax(smatrix))
cbar = plt.colorbar(heatmap, shrink=0.9, pad=0.01)
cbar.set_label(xtitle, rotation=270, labelpad=30)
cbar.ax.tick_params(labelsize=10)
ticks = [n - 0.5 for n in range(1, len(xticklabels) + 1, 1)]
# xticklabels = [str(int(n + 0.5)) for n in ticks]
yticklabels = xticklabels
# if n_biggest_clusters > 20:
# modulo = 3
# ticks = [ticks[it] for it in range(0, len(ticks), modulo)]
# xticklabels = [b_cluster_lengths[it] for it in range(0, len(b_cluster_lengths), modulo)]
# yticklabels = [a_cluster_lengths[it] for it in range(0, len(a_cluster_lengths), modulo)]
plt.xticks(ticks, xticklabels, rotation=90)
plt.yticks(ticks, yticklabels)
# plt.xlabel(legends.get(meth2, meth2) + ' cluster size') # I don't know why it's reversed, it just is
# plt.ylabel(legends.get(meth1, meth1) + ' cluster size')
# ax.set_xlim(0, n_biggest_clusters)
# ax.set_ylim(0, n_biggest_clusters)
plt.title(title)
if not os.path.exists(plotdir + '/plots'):
os.makedirs(plotdir + '/plots')
plt.savefig(plotdir + '/plots/' + plotname + '.svg')
plt.close()
def plotheatmap(plotdir, plotname, difftype, genelist=None, genesets=None, title='', xtitle=''):
assert genelist is None or genesets is None
if genelist is not None:
smatrix = get_gene_pair_matrix(genelist, difftype)
xticklabels = [utils.summarize_gene_name(g) for g in genelist]
elif genesets is not None:
smatrix = get_gene_set_mean_matrix(genesets, difftype)
xticklabels = [sn for sn in genesets.keys()]
else:
raise Exception('no gene list specified')
assert len(smatrix) == len(smatrix[0]) # uh, I think I need this to be true
fig, ax = plotting.mpl_init()
plt.tick_params(axis='both', which='major', labelsize=7)
plt.gcf().subplots_adjust(bottom=0.14, left=0.18, right=0.95, top=0.92)
ax.xaxis.set_ticks_position('bottom')
ax.yaxis.set_ticks_position('left')
data = numpy.array(smatrix)
cmap = plt.cm.Blues #cm.get_cmap('jet')
cmap.set_under('w')
heatmap = ax.pcolor(data, cmap=cmap, vmin=0., vmax=0.5) #vmax=numpy.amax(smatrix))
cbar = plt.colorbar(heatmap, shrink=0.9, pad=0.01)
cbar.set_label(xtitle, rotation=270, labelpad=30)
cbar.ax.tick_params(labelsize=10)
ticks = [n - 0.5 for n in range(1, len(xticklabels) + 1, 1)]
# xticklabels = [str(int(n + 0.5)) for n in ticks]
yticklabels = xticklabels
# if n_biggest_clusters > 20:
# modulo = 3
# ticks = [ticks[it] for it in range(0, len(ticks), modulo)]
# xticklabels = [b_cluster_lengths[it] for it in range(0, len(b_cluster_lengths), modulo)]
# yticklabels = [a_cluster_lengths[it] for it in range(0, len(a_cluster_lengths), modulo)]
plt.xticks(ticks, xticklabels, rotation=90)
plt.yticks(ticks, yticklabels)
# plt.xlabel(legends.get(meth2, meth2) + ' cluster size') # I don't know why it's reversed, it just is
# plt.ylabel(legends.get(meth1, meth1) + ' cluster size')
# ax.set_xlim(0, n_biggest_clusters)
# ax.set_ylim(0, n_biggest_clusters)
plt.title(title)
if not os.path.exists(plotdir + '/plots'):
os.makedirs(plotdir + '/plots')
plt.savefig(plotdir + '/plots/' + plotname + '.svg')
plt.close()
def make_mutefreq_plot(plotdir, gene_name, positions):
import plotting
""" NOTE shares a lot with make_transition_plot() in bin/plot-hmms.py. """
nuke_colors = {'A' : 'red', 'C' : 'blue', 'G' : 'orange', 'T' : 'green'}
fig, ax = plotting.mpl_init()
fig.set_size_inches(plotting.plot_ratios[utils.get_region(gene_name)])
ibin = 0
print utils.color_gene(utils.unsanitize_name(gene_name))
legend_colors = set()
for info in positions:
posname = info['name']
# make label below bin
ax.text(-0.5 + ibin, -0.075, simplify_state_name(posname), rotation='vertical', size=8)
total = 0.0
alpha = 0.6
for nuke, prob in sorted(info['nuke_freqs'].items(), key=operator.itemgetter(1), reverse=True):
color = nuke_colors[nuke]
label_to_use = None
if color not in legend_colors:
label_to_use = nuke
legend_colors.add(color)
# horizontal line at height total+prob
ax.plot([-0.5 + ibin, 0.5 + ibin], [total + prob, total + prob], color=color, alpha=alpha, linewidth=3, label=label_to_use)
# vertical line from total to total + prob
ax.plot([ibin, ibin], [total + 0.01, total + prob], color=color, alpha=alpha, linewidth=3)
# # write [ACGT] at midpoint between total and total+prob
# midpoint = 0.5*(prob + 2*total)
# ... *redacted*
total += prob
ibin += 1
ax.get_xaxis().set_visible(False)
plotting.mpl_finish(ax, plotdir, gene_name, ybounds=(-0.01, 1.01), xbounds=(-3, len(positions) + 3), leg_loc=(0.95, 0.1), adjust={'left' : 0.1, 'right' : 0.8}, leg_prop={'size' : 8})
def make_mutefreq_plot(plotdir, gene_name, positions):
""" NOTE shares a lot with make_transition_plot() in bin/plot-hmms.py. """
nuke_colors = {'A' : 'red', 'C' : 'blue', 'G' : 'orange', 'T' : 'green'}
fig, ax = plotting.mpl_init()
fig.set_size_inches(plotting.plot_ratios[utils.get_region(gene_name)])
ibin = 0
print utils.color_gene(utils.unsanitize_name(gene_name))
legend_colors = set()
for info in positions:
posname = info['name']
# make label below bin
ax.text(-0.5 + ibin, -0.075, simplify_state_name(posname), rotation='vertical', size=8)
total = 0.0
alpha = 0.6
for nuke, prob in sorted(info['nuke_freqs'].items(), key=operator.itemgetter(1), reverse=True):
color = nuke_colors[nuke]
label_to_use = None
if color not in legend_colors:
label_to_use = nuke
legend_colors.add(color)
# horizontal line at height total+prob
ax.plot([-0.5 + ibin, 0.5 + ibin], [total + prob, total + prob], color=color, alpha=alpha, linewidth=3, label=label_to_use)
# vertical line from total to total + prob
ax.plot([ibin, ibin], [total + 0.01, total + prob], color=color, alpha=alpha, linewidth=3)
# # write [ACGT] at midpoint between total and total+prob
# midpoint = 0.5*(prob + 2*total)
# ... *redacted*
total += prob
ibin += 1
ax.get_xaxis().set_visible(False)
plotting.mpl_finish(ax, plotdir, gene_name, ybounds=(-0.01, 1.01), xbounds=(-3, len(positions) + 3), leg_loc=(0.95, 0.1), adjust={'left' : 0.1, 'right' : 0.8}, leg_prop={'size' : 8})
def peruse_naive_seqs():
from hist import Hist
# hall = Hist(n_set_list[-1], n_set_list[0] - 0.5, n_set_list[-1] + 0.5)
means = []
for n_set in n_set_list:
plotdir = baseplotdir + '/' + str(n_set)
hist = Hist(fname=plotdir + '/hmm/hamming_to_true_naive.csv')
print '%2d %.2f' % (n_set, hist.get_mean()),
# hall.set_ibin(hall.find_bin(n_set), hist.get_mean())
means.append(hist.get_mean())
import plotting
fig, ax = plotting.mpl_init()
# hall.mpl_plot(ax)
ax.plot(n_set_list, means, marker='.')
plotting.mpl_finish(ax,
baseplotdir,
'means',
xlabel='N simultaneous seqs',
ylabel='mean hamming to true naive',
ybounds=(0, None))
def make_single_size_vs_shm_plot(self,
sorted_clusters,
annotations,
repertoire_size,
base_plotdir,
plotname,
n_max_mutations=100,
plot_high_mutation=False,
title=None,
debug=False):
import plotting
def gety(minval, maxval, xmax, x):
slope = (maxval - minval) / xmax
return slope * x + minval
def getnmutelist(cluster):
return annotations[':'.join(cluster)]['n_mutations']
colors = ['#006600', '#3399ff', '#ffa500']
# goldenrod '#daa520'
# red '#cc0000',
# dark red '#990012'
# purple '#a821c7'
# grey '#808080'
dpi = 80
xpixels = 450
ypixels = max(400, 10 * len(sorted_clusters))
fig, ax = plotting.mpl_init(figsize=(xpixels / dpi, ypixels / dpi))
min_linewidth = 0.3
max_linewidth = 12
# min_alpha = 0.1
# max_alpha = 1.
# linewidth = 7
alpha = 0.55
ymin, ymax = 9999, 0
iclust_global = 0
yticks, yticklabels = [], []
high_mutation_clusters = []
biggest_n_mutations = None
if debug:
print ' %s %d x %d' % (
plotname, xpixels, ypixels
) #, utils.color('red', 'high mutation') if plot_high_mutation else '')
print ' size frac yval median mean'
for csize, cluster_group in itertools.groupby(sorted_clusters,
key=lambda c: len(c)):
cluster_group = sorted(list(cluster_group),
key=lambda c: numpy.median(getnmutelist(c)))
n_clusters = len(cluster_group)
repfracstr = self.get_repfracstr(csize, repertoire_size)
for iclust in range(len(cluster_group)):
cluster = cluster_group[iclust]
nmutelist = sorted(getnmutelist(cluster))
nmedian = numpy.median(nmutelist)
nmean = numpy.mean(
nmutelist) # maybe should use this instead of the median?
if biggest_n_mutations is None or nmutelist[
-1] > biggest_n_mutations:
biggest_n_mutations = nmutelist[-1]
yval = len(sorted_clusters) - iclust_global
if yval < ymin:
ymin = yval
if yval > ymax:
ymax = yval
yticks.append(yval)
yticklabels.append('%d' % csize)
# yticklabels.append(repfracstr)
base_color = colors[iclust_global % len(colors)]
if self.args.queries_to_include is not None:
queries_to_include_in_this_cluster = set(cluster) & set(
self.args.queries_to_include)
if len(queries_to_include_in_this_cluster) > 0:
base_color = 'red'
if plot_high_mutation:
xtext = 1.1
elif float(nmedian) / n_max_mutations < 0.5:
xtext = 0.75
else:
xtext = 0.1
ax.text(xtext * n_max_mutations,
yval,
' '.join(queries_to_include_in_this_cluster),
color='red',
fontsize=8)
if debug:
print ' %5s %-10s %4.1f %6.1f %6.1f' % (
'%d' % csize if iclust == 0 else '', repfracstr
if iclust == 0 else '', yval, nmedian, nmean),
if nmedian > n_max_mutations and not plot_high_mutation:
if debug:
print '%s' % utils.color('red', 'high mutation')
high_mutation_clusters.append(cluster)
continue
if debug:
print ''
nbins = nmutelist[-1] - nmutelist[0] + 1
hist = Hist(nbins, nmutelist[0] - 0.5, nmutelist[-1] + 0.5)
for nm in nmutelist:
hist.fill(nm)
assert hist.overflow_contents() == 0. # includes underflows
xmax = max(hist.bin_contents) # float(csize)
for ibin in range(1, hist.n_bins + 1):
linewidth = gety(min_linewidth, max_linewidth, xmax,
hist.bin_contents[ibin])
color = base_color
# alpha = gety(min_alpha, max_alpha, xmax, hist.bin_contents[ibin])
if hist.bin_contents[ibin] == 0.:
color = 'grey'
linewidth = min_linewidth
alpha = 0.4
ax.plot([hist.low_edges[ibin], hist.low_edges[ibin + 1]],
[yval, yval],
color=color,
linewidth=linewidth,
alpha=alpha,
solid_capstyle='butt')
iclust_global += 1
xbounds = [-0.2, n_max_mutations] if not plot_high_mutation else [
n_max_mutations, biggest_n_mutations
]
ybounds = [0.95 * ymin, 1.05 * ymax]
n_ticks = 5
if len(yticks) > n_ticks:
yticks = [
yticks[i] for i in range(0, len(yticks),
int(len(yticks) / float(n_ticks - 1)))
]
yticklabels = [
yticklabels[i]
for i in range(0, len(yticklabels),
int(len(yticklabels) / float(n_ticks - 1)))
]
plotting.mpl_finish(ax,
base_plotdir + '/overall',
plotname,
xlabel='N mutations',
ylabel='clonal family size',
title=title,
xbounds=xbounds,
ybounds=ybounds,
yticks=yticks,
yticklabels=yticklabels,
adjust={'left': 0.18})
return high_mutation_clusters
def fullplot(self, plotdir, plotname, **kwargs): # i.e. full plotting process, not just the ax.plot type stuff above
import plotting
fig, ax = plotting.mpl_init() # this'll need to be updated when i want to use a kwarg for this fcn
self.mpl_plot(ax)
plotting.mpl_finish(ax, plotdir, plotname, **kwargs)
self.write('%s/%s.csv'%(plotdir, plotname))
def peruse_forward_scores():
_, reco_info = seqfileopener.get_seqfile_info(
simfname, is_data=False) #, n_max_queries=10000)
logprobs, partialcorr_logprobs, corr_logprobs = OrderedDict(), OrderedDict(
), OrderedDict()
for n_set in n_set_list:
print n_set
# if n_set != 5:
# continue
logprobs[n_set], partialcorr_logprobs[n_set], corr_logprobs[
n_set] = OrderedDict(), OrderedDict(), OrderedDict()
with open(outputdir + '/' + str(n_set) + '-forward.csv') as csvfile:
reader = csv.DictReader(csvfile)
for line in reader:
uidlist = line['unique_ids'].split(':')
assert utils.from_same_event(reco_info, uidlist)
reco_id = reco_info[uidlist[0]]['reco_id']
if reco_id in logprobs[n_set]:
raise Exception('already had %s' % reco_id)
logprobs[n_set][reco_id] = float(line['logprob'])
factor = 1. / n_set
partialcorr_logprobs[n_set][reco_id] = factor * float(
line['logprob'])
factor = (1. - 0.24 / pow(float(n_set), 0.9)) / n_set
# factor = 1. / (0.77547824*n_set + 0.20327936)
corr_logprobs[n_set][reco_id] = factor * float(line['logprob'])
i_baseline = -1
deviations = get_deviations(logprobs, i_baseline)
# fit_stuff(n_set_list, deviations)
partialcorr_deviations = get_deviations(partialcorr_logprobs, i_baseline)
signed_partialcorr_deviations = get_deviations(partialcorr_logprobs,
i_baseline,
signed=True)
corr_deviations = get_deviations(corr_logprobs, i_baseline)
signed_corr_deviations = get_deviations(corr_logprobs,
i_baseline,
signed=True)
import plotting
fig, ax = plotting.mpl_init()
ax.plot(n_set_list, deviations, marker='.')
plotting.mpl_finish(ax,
baseplotdir,
'forwards',
xlabel='N simultaneous seqs',
ylabel='log prob deviation to ' +
str(n_set_list[i_baseline])) #, ybounds=(-0.02, 0.02))
# fig, ax = plotting.mpl_init()
# ax.plot(n_set_list, partialcorr_deviations, marker='.')
# ax.plot([n_set_list[0], n_set_list[-1]], [0, 0])
# plotting.mpl_finish(ax, baseplotdir, 'partially-corrected-forwards', xlabel='N simultaneous seqs', ylabel='log prob deviation to ' + str(n_set_list[i_baseline])) #, ybounds=(-0.02, 0.02))
fig, ax = plotting.mpl_init()
ax.plot(n_set_list, partialcorr_deviations, marker='.', label='1/n (abs)')
ax.plot(n_set_list, signed_partialcorr_deviations, marker='.', label='1/n')
ax.plot(n_set_list, corr_deviations, marker='.', label='1/crap (abs)')
ax.plot(n_set_list, signed_corr_deviations, marker='.', label='1/crap')
ax.plot([n_set_list[0], n_set_list[-1]], [0, 0])
plotting.mpl_finish(ax,
baseplotdir,
'corrected-forwards',
xlabel='N simultaneous seqs',
ylabel='log prob deviation to ' +
str(n_set_list[i_baseline])) #, ybounds=(-0.02, 0.02))
fig, ax = plotting.mpl_init()
ax.plot(n_set_list, signed_corr_deviations, marker='.')
ax.plot([n_set_list[0], n_set_list[-1]], [0, 0])
plotting.mpl_finish(ax,
baseplotdir,
'signed-corrected-forwards',
xlabel='N simultaneous seqs',
ylabel='log prob deviation to ' +
str(n_set_list[i_baseline])) #, ybounds=(-0.02, 0.02))
chimeric_fraction = n_above_cutoff / float(len(chfo))
print ' %d / %d = %.3f above chimeric cutoff' % (n_above_cutoff, len(chfo),
chimeric_fraction)
hmaxval = Hist(45, 0., 0.65)
for uid in annotations:
hmaxval.fill(chfo[uid]['max_abs_diff'])
himax = Hist(75, 0., 400)
for uid in annotations:
himax.fill(chfo[uid]['imax'])
utils.prep_dir(args.plotdir, wildlings=['*.svg', '*.csv'])
import matplotlib
from matplotlib import pyplot as plt
fig, ax = plotting.mpl_init()
xvals, yvals = zip(*[(v['imax'], v['max_abs_diff']) for v in chfo.values()])
plt.scatter(xvals, yvals, alpha=0.4)
print 'writing to %s' % args.plotdir
plotting.mpl_finish(ax,
args.plotdir,
'hexbin',
title=args.title,
xlabel='break point',
ylabel='abs mfreq diff')
plotting.draw_no_root(hmaxval,
plotdir=args.plotdir,
plotname='mfreq-diff',
shift_overflows=True,
def plot(self, plotdir, only_csv=False):
utils.prep_dir(plotdir, wildling=None, multilings=['*.csv', '*.svg', '*.root'])
for column in self.values:
if self.only_correct_gene_fractions and column not in bool_columns:
continue
if column in bool_columns:
right = self.values[column]['right']
wrong = self.values[column]['wrong']
errs = fraction_uncertainty.err(right, right+wrong)
print ' %s\n correct up to allele: %4d / %-4d = %4.4f (-%.3f, +%.3f)' % (column, right, right+wrong, float(right) / (right + wrong), errs[0], errs[1])
hist = plotting.make_bool_hist(right, wrong, self.name + '-' + column)
plotting.draw_no_root(hist, plotname=column, plotdir=plotdir, write_csv=True, stats='0-bin', only_csv=only_csv)
else:
# TODO this is dumb... I should make the integer-valued ones histograms as well
hist = plotting.make_hist_from_dict_of_counts(self.values[column], 'int', self.name + '-' + column, normalize=True)
log = ''
if column.find('hamming_to_true_naive') >= 0: # TODO why doesn't this just use the config dicts in plotheaders or wherever?
hist.title = 'hamming distance'
else:
hist.title = 'inferred - true'
plotting.draw_no_root(hist, plotname=column, plotdir=plotdir, write_csv=True, log=log, only_csv=only_csv)
for column in self.hists:
plotting.draw_no_root(self.hists[column], plotname=column, plotdir=plotdir, write_csv=True, log=log, only_csv=only_csv)
# per-gene support crap
for region in utils.regions:
if self.hists[region + '_allele_right_vs_per_gene_support'].integral(include_overflows=True) == 0:
continue
xvals = self.hists[region + '_allele_right_vs_per_gene_support'].get_bin_centers() #ignore_overflows=True)
right = self.hists[region + '_allele_right_vs_per_gene_support'].bin_contents
wrong = self.hists[region + '_allele_wrong_vs_per_gene_support'].bin_contents
yvals = [float(r) / (r + w) if r + w > 0. else 0. for r, w in zip(right, wrong)]
# remove values corresponding to bins with no entries
while yvals.count(0.) > 0:
iv = yvals.index(0.)
xvals.pop(iv)
right.pop(iv)
wrong.pop(iv)
yvals.pop(iv)
tmphilos = [fraction_uncertainty.err(r, r + w) for r, w in zip(right, wrong)]
yerrs = [err[1] - err[0] for err in tmphilos]
# fitting a line isn't particularly informative, actually
# params, cov = numpy.polyfit(xvals, yvals, 1, w=[1./(e*e) if e > 0. else 0. for e in yerrs], cov=True)
# slope, slope_err = params[0], math.sqrt(cov[0][0])
# y_icpt, y_icpt_err = params[1], math.sqrt(cov[1][1])
# print '%s slope: %5.2f +/- %5.2f y-intercept: %5.2f +/- %5.2f' % (region, slope, slope_err, y_icpt, y_icpt_err)
# print '%s' % region
# for iv in range(len(xvals)):
# print ' %5.2f %5.0f / %5.0f = %5.2f +/- %.3f' % (xvals[iv], right[iv], right[iv] + wrong[iv], yvals[iv], yerrs[iv])
fig, ax = plotting.mpl_init()
ax.errorbar(xvals, yvals, yerr=yerrs, markersize=10, linewidth=1, marker='.')
ax.plot((0, 1), (0, 1), color='black', linestyle='--', linewidth=3) # line with slope 1 and intercept 0
# linevals = [slope*x + y_icpt for x in [0] + xvals] # fitted line
# ax.plot([0] + xvals, linevals)
plotting.mpl_finish(ax, plotdir, region + '_allele_fraction_correct_vs_per_gene_support', xlabel='support', ylabel='fraction correct', xbounds=(-0.1, 1.1), ybounds=(-0.1, 1.1))
if not only_csv:
plotting.make_html(plotdir)
def make_mutefreq_plot(plotdir, gene_name, positions, debug=False):
import plotting
""" NOTE shares a lot with make_transition_plot() in bin/plot-hmms.py. """
nuke_colors = {'A': 'red', 'C': 'blue', 'G': 'orange', 'T': 'green'}
fig, ax = plotting.mpl_init()
fig.set_size_inches(plotting.plot_ratios[utils.get_region(gene_name)])
ibin = 0
if debug:
print ' %s' % utils.color_gene(utils.unsanitize_name(gene_name))
legend_colors = set()
for info in positions:
posname = info['name']
# make label below bin for position and germline nuke
ax.text(-0.5 + ibin,
-0.075,
simplify_state_name(posname),
rotation='vertical',
size=8)
ax.text(-0.5 + ibin,
-0.15,
info.get('gl_nuke', '?'),
fontsize=10,
fontweight='bold')
sorted_nukes, _ = zip(*sorted(info['nuke_freqs'].items(),
key=operator.itemgetter(1),
reverse=True))
if 'gl_nuke' in info and info['gl_nuke'] in info[
'nuke_freqs']: # put the germline nuke first if we have it (second clause is for states with germline N))
sorted_nukes = [info['gl_nuke']] + [
n for n in sorted_nukes if n != info['gl_nuke']
]
total = 0.0
alpha = 0.6
for nuke in sorted_nukes:
prob = info['nuke_freqs'][nuke]
color = nuke_colors[nuke]
label_to_use = None
if color not in legend_colors:
label_to_use = nuke
legend_colors.add(color)
# horizontal line at height total+prob
ax.plot([-0.5 + ibin, 0.5 + ibin], [total + prob, total + prob],
color=color,
alpha=alpha,
linewidth=3,
label=label_to_use)
# vertical line from total to total + prob
ax.plot([ibin, ibin], [total + 0.01, total + prob],
color=color,
alpha=alpha,
linewidth=3)
# # write [ACGT] at midpoint between total and total+prob
# midpoint = 0.5*(prob + 2*total)
# ... *redacted*
total += prob
ibin += 1
ax.get_xaxis().set_visible(False)
plotting.mpl_finish(ax,
plotdir,
gene_name,
ybounds=(-0.01, 1.01),
xbounds=(-3, len(positions) + 3),
leg_loc=(0.95, 0.1),
adjust={
'left': 0.1,
'right': 0.8
},
leg_prop={'size': 8})
def make_single_hexbin_size_vs_shm_plot(self,
sorted_clusters,
annotations,
repertoire_size,
base_plotdir,
plotname,
n_max_mutations=100,
log_cluster_size=False,
debug=False):
import plotting
import matplotlib.pyplot as plt
def getnmutelist(cluster):
return annotations[':'.join(cluster)]['n_mutations']
fig, ax = plotting.mpl_init()
xvals, yvals = zip(
*[[numpy.mean(getnmutelist(cluster)),
len(cluster)] for cluster in sorted_clusters
if numpy.mean(getnmutelist(cluster)) < n_max_mutations])
if log_cluster_size:
yvals = [math.log(yv) for yv in yvals]
hb = ax.hexbin(xvals,
yvals,
gridsize=n_max_mutations,
cmap=plt.cm.Blues,
bins='log')
nticks = 5
yticks = [
yvals[0] + itick * (yvals[-1] - yvals[0]) / float(nticks - 1)
for itick in range(nticks)
]
if log_cluster_size:
yticklabels = [math.exp(yt) for yt in yticks]
yticklabels = [('%.0f' % yt) if yt > 5 else ('%.1f' % yt)
for yt in yticklabels]
else:
yticklabels = [int(yt) for yt in yticks]
if self.args.queries_to_include is not None:
for cluster in sorted_clusters:
queries_to_include_in_this_cluster = set(cluster) & set(
self.args.queries_to_include)
if len(queries_to_include_in_this_cluster) == 0:
continue
xval = numpy.mean(getnmutelist(cluster))
yval = len(cluster)
if log_cluster_size:
yval = math.log(yval)
ax.plot([xval], [yval], color='red', marker='.', markersize=10)
ax.text(xval,
yval,
' '.join(queries_to_include_in_this_cluster),
color='red',
fontsize=8)
ylabel = 'clonal family size'
if log_cluster_size:
ylabel += ' (log)'
plotname += '-log'
plotting.mpl_finish(ax,
base_plotdir + '/overall',
plotname,
xlabel='mean N mutations',
ylabel=ylabel,
xbounds=[0, n_max_mutations],
yticks=yticks,
yticklabels=yticklabels)
