def scatter(files, tracksheets, ttypes, colors):
nuceriod_plt.config_params(12)
df = _load(files)
toplot = df.merge(tracksheets, how='left')
toplot['ttype'] = toplot['tumor_name'].map(ttypes)
fig, ax = plt.subplots(nrows=1, ncols=1, figsize=(16, 4))
ax.set_xscale('log')
for i, row in toplot[(toplot['qvals_snr'] < 0.1)
& (toplot['snr'] > 8)].iterrows():
ax.scatter(row['muts'],
row['prop_increase_in'],
s=14,
color=colors[row['ttype']])
plt.title('Significant')
plt.ylabel('Proportion of increase minor in')
plt.xlabel('Number of mutations (log)')
plt.hlines(0, 0, 1000000, linestyle='--', color='grey', alpha=0.6)
ax.spines['top'].set_visible(False)
ax.spines['right'].set_visible(False)
plt.ylim(-0.25, 0.25)
def rotational_bars(rot_high_files, rot_low_files):
nuceriod_plt.config_params(11)
df_high = increase.load_d(rot_high_files)
df_high['rot'] = 'high'
df_low = increase.load_d(rot_low_files)
df_low['rot'] = 'low'
df = pd.concat([df_high, df_low])
fig, axs = plt.subplots(nrows=len(df.groupby(by='name')),
ncols=1,
figsize=(1.75, 8))
order = ['low', 'high']
for ix, (sig, data) in enumerate(df.groupby(by='name')):
xvals = []
yvals = []
colors = []
count = 0
for i in order:
val = data[data['rot'] == i]['snr'].tolist()[0]
yvals.append(val)
xvals.append(count)
colors.append(COLORS[sig])
count += 1
axs[ix].bar(xvals, yvals, color=colors, label=['low', 'high'])
axs[ix].set_xticks([0, 1])
axs[ix].set_xticklabels(('low', 'high'), fontsize=11)
axs[ix].set_ylabel('SNR')
axs[ix].spines['right'].set_visible(False)
axs[ix].spines['top'].set_visible(False)
plt.tight_layout()
def compare(files_deconstructsigs, files_sigfit):
nuceriod_plt.config_params(11)
df_deconstructsigs = increase.load_d(files_deconstructsigs)
df_deconstructsigs['control'] = 'deconstructsigs'
df_sigfit = increase.load_d(files_sigfit)
df_sigfit['control'] = 'sigfit'
toplot = pd.concat([df_deconstructsigs, df_sigfit])
fig, ax = plt.subplots(nrows=1, ncols=1, figsize=(3, 2))
sig_list = []
decon_list = []
sig_list2 = []
decon_list2 = []
for i, data in toplot.groupby(by='name'):
if len(data) == 2:
row = data.iloc[0]
snr_decon = data[data['control'] == 'sigfit']['snr'].iloc[0]
snr_sigfit = data[data['control'] ==
'deconstructsigs']['snr'].iloc[0]
sig_list.append(snr_sigfit)
decon_list.append(snr_decon)
if row['cross_validation_max'] < 0:
ax.scatter(-np.log(snr_decon),
-np.log(snr_sigfit),
c=COLORS[i])
sig_list2.append(-np.log(snr_sigfit))
decon_list2.append(-np.log(snr_decon))
else:
ax.scatter(np.log(snr_decon), np.log(snr_sigfit), c=COLORS[i])
sig_list2.append(np.log(snr_sigfit))
decon_list2.append(np.log(snr_decon))
plt.xlabel('Period')
ylabels = [str(2**abs(i)) for i in range(2, 10, 2)]
yfinal = ylabels[::-1] + ylabels[1:]
plt.xticks(np.arange(-6, 8, 2), yfinal)
ylabels = [str(2**abs(i)) for i in range(2, 10, 2)]
yfinal = ylabels[::-1] + ylabels[1:]
plt.yticks(np.arange(-6, 8, 2), yfinal)
slope, intercept, r_value, p_value, std_err = stats.linregress(
sig_list2, decon_list2)
xvals = np.arange(-6, 8, 2)
yvals = [slope * y + intercept for y in xvals]
plt.plot(xvals, yvals)
R, pval = stats.pearsonr(sig_list, decon_list)
plt.text(-4, 3, 'R = {}\npval = {}'.format(round(R, 3), round(pval, 3)))
ax.spines['top'].set_visible(False)
ax.spines['right'].set_visible(False)
plt.ylabel('SNR (Sigfit)')
plt.xlabel('SNR (Deconstructsig)')
plt.tight_layout()
def autocorrelation(original_file, simulated_files):
nucperiod_plt.config_params(font_size=14)
with open(original_file, 'rt') as f:
mydict = json.load(f)
pair_count_array = np.array(mydict['pair_count'])
motif_count = mydict['motif_count']
len_chunk = mydict['chunk_len']
# define the signal
signal = np.array([(v / len_chunk) / (motif_count / len_chunk) ** 2 for v in pair_count_array])
# define figure
figsize = (10, 5)
fig, ax = plt.subplots(1, 1, figsize=figsize)
ax.set_title('motif autocorrelation')
# raw signal
ax.plot(range(4, len(signal)), signal[4:], alpha=0.5, label='raw')
# 3-smoothing
signal = spectral.mean_smooth(signal, 3)
ax.plot(range(4, len(signal)), signal[4:], linewidth=3.0, label='3-smoothed')
counter = 0
# detrended smooth autocorrelation
initial_values_dict = {'a_0': np.mean(signal[4:]), 'a_1': 0., 'a_2': 0.}
params, obj_func = non_linear.create_quadratic_model(initial_values_dict)
x = np.arange(len(signal[4:]))
non_linear_fitter = non_linear.NonLinearFitting(obj_func, params, x, signal[4:])
_, predicted = non_linear_fitter.least_squares()
# with quadratic least-squares fit
ax.plot(range(4, len(signal)), predicted, '--', label='quadratic trend')
for file in simulated_files:
counter += 1
with open(file, 'rt') as f:
random_chunk = json.load(f)
pc = np.array(random_chunk['pair_count'])
mc = random_chunk['motif_count']
len_random_chunk = random_chunk['chunk_len']
random_signal = np.array([(v / len_chunk) / (mc / len_random_chunk) ** 2 for v in pc])
random_signal = spectral.mean_smooth(random_signal, 3)
if counter == 1:
label = 'randomized'
else:
label = None
ax.plot(range(4, len(random_signal)), random_signal[4:], linewidth=3.0,
alpha=0.3, color='grey', label=label)
ax.legend()
plt.rcParams['savefig.facecolor'] = fig.get_facecolor()
def compare(cohorts_5mer, cohorts_3mer, cohorts_linker, tumors=None):
nuceriod_plt.config_params(14)
df_5mer = increase.load_d(cohorts_5mer)
df_5mer['control'] = 'mer5'
df_3mer = increase.load_d(cohorts_3mer)
df_3mer['control'] = 'mer3'
df_linker = increase.load_d(cohorts_linker)
df_linker['control'] = 'linker'
df = pd.concat([df_5mer, df_3mer, df_linker])
df['increase_in'] = df['observed_in'] - df['expected_in']
df['prop_increase_in'] = df['increase_in'] / df['expected_in']
df['ttype'] = df['name'].map(TTYPES)
if tumors is not None:
df = df[df['name'].isin(tumors)]
toplot = df.sort_values(by='prop_increase_in', ascending=True)
fig, ax = plt.subplots(nrows=1, ncols=1, figsize=(10, 5))
labels = []
colors = []
red_patch = mpatches.Patch(color='red', label='3-mers')
green_patch = mpatches.Patch(color='green',
label='no nucleosomes in context')
orange_patch = mpatches.Patch(color='orange', label='5-mers')
for ix, (ttype, data) in enumerate(
toplot.sort_values(by='snr', ascending=True).groupby(by='ttype',
sort=False)):
snr1 = data[data['control'] == 'mer3']['snr'].tolist()[0]
snr2 = data[data['control'] == 'linker']['snr'].tolist()[0]
snr3 = data[data['control'] == 'mer5']['snr'].tolist()[0]
colors.append('red')
colors.append('green')
colors.append('orange')
ax.scatter(ix, math.log2(snr1), color='red', s=15, alpha=0.8)
ax.scatter(ix, math.log2(snr2), color='green', s=15, alpha=0.8)
ax.scatter(ix, math.log2(snr3), color='orange', s=15, alpha=0.8)
labels.append(ttype)
plt.xticks([i for i in range(ix + 1)], labels, rotation=90)
tick = [2, 4, 6, 8]
plt.yticks(tick, [str(2**t) for t in tick])
plt.ylabel('log2(SNR)')
plt.legend(handles=[red_patch, green_patch, orange_patch])
ax.spines['top'].set_visible(False)
ax.spines['right'].set_visible(False)
plt.tight_layout()
def plot_bars(snr_high, snr_low):
nucperiod_plt.config_params()
fig, axs = plt.subplots(nrows=1, ncols=1, figsize=(0.9,1.1,))
xvals = [0, 1]
yvals = [snr_low, snr_high]
axs.bar(xvals, yvals, label=['low','high'], color=['#afdde9ff', '#afdde9ff'])
plt.xticks(xvals, ['low', 'high'], fontsize=7)
axs.set_xlabel('SNR', fontsize = 7)
axs.spines['right'].set_visible(False)
axs.spines['top'].set_visible(False)
plt.tight_layout()
def rotational(cohorts_high, cohorts_low, tumors=None):
nuceriod_plt.config_params(14)
df_high = increase.load_d(cohorts_high)
df_high['control'] = 'high'
df_low = increase.load_d(cohorts_low)
df_low['control'] = 'low'
df = pd.concat([df_high, df_low])
df['increase_in'] = df['observed_in'] - df['expected_in']
df['prop_increase_in'] = df['increase_in'] / df['expected_in']
df['ttype'] = df['name'].map(TTYPES)
if tumors is not None:
df = df[df['name'].isin(tumors)]
toplot = df.sort_values(by='prop_increase_in', ascending=True)
order = ['low', 'high']
count = 0
xvals = []
yvals = []
colors = []
dic_t = collections.defaultdict(dict)
labels = []
for sig, data in toplot.sort_values(by='snr').groupby(by='ttype',
sort=False):
if sig in COLORS:
labels.append('{}'.format(sig))
for i in order:
val = data[data['control'] == i]['snr'].tolist()[0]
dic_t['Sign {}'.format(sig)][i] = val
yvals.append(math.log2(val))
xvals.append(count)
colors.append(COLORS[sig])
count += 1
fig, ax = plt.subplots(nrows=1, ncols=1, figsize=(10, 1.5))
ax.bar(xvals, yvals, color=colors, label=labels)
ax.set_ylabel('SNR')
plt.xticks(np.arange(0.5, 54, 2), labels, rotation=90, fontsize=13)
tick = [2, 4, 6, 8]
plt.yticks(tick, [str(2**t) for t in tick])
ax.spines['top'].set_visible(False)
ax.spines['right'].set_visible(False)
def sapiens_bars(snr_high, snr_medium, snr_low):
nucperiod_plt.config_params(7)
fig, axs = plt.subplots(nrows=1, ncols=1, figsize=(1.3, 1.5))
xvals = [0, 1, 2]
yvals = [snr_low, snr_medium, snr_high]
axs.bar(xvals,
yvals,
label=['low', 'medium', 'high'],
color=['#afdde9ff', '#afdde9ff', '#afdde9ff'])
axs.set_xticks([0, 1, 2])
axs.set_xticklabels(('very-low', 'low', 'high'), rotation=90)
axs.set_ylabel('SNR')
axs.spines['right'].set_visible(False)
axs.spines['top'].set_visible(False)
plt.tight_layout()
def scatter(df, feature1, feature2, selected_organisms=None, **kwargs):
just_once = True
if selected_organisms is None:
selected_organisms = []
nucperiod_plt.config_params(font_size=24)
fig, axes = plt.subplots(1, 1, figsize=(12, 14))
phyllum = ['protists', 'fungi', 'plants', 'vertebrates', 'insects', 'nematodes', 'deuterostomes']
color_label = dict(zip(phyllum, ['blue', 'grey', 'green', 'pink', 'black', 'cyan', 'orange']))
for org_type in phyllum:
ds = df[df['org_type'] == org_type]
linewidths_normal = [1 for _ in ds[feature1].values]
linewidths_snr = [2 if a < 1e-2 and b > 0 else 0. for a, b in zip(ds['qval_power_enrichment'].values, ds[feature1].values)]
if just_once:
axes.scatter([], [], s=300, linewidths=linewidths_snr,
edgecolor='#8b0000ff', color='None', label='q-value < 0.01')
just_once = False
axes.scatter(ds[feature1].values, ds[feature2].values, s=350, linewidths=linewidths_snr,
edgecolor='#8b0000ff', color='None')
axes.scatter(ds[feature1].values, ds[feature2].values, s=150, linewidths=linewidths_normal,
edgecolor='black', color=color_label[org_type], label=org_type)
for i, txt in enumerate(df.index.values):
if txt in selected_organisms:
axes.annotate(txt, (df.loc[txt, feature1] + 0.02, df.loc[txt, feature2] + 0.02))
axes.set_ylabel('Proportion of 1 Mb chunks with with MP at {0} $\pm$ 0.5 bp (ratio)'.format(kwargs['period']))
axes.set_xlabel('Power Enrichment at {0} $\pm$ 0.5 bp (odds ratio)'.format(kwargs['period']))
# Power enrichment: power significance at {0} $\pm$ 0.5 bp compared to other periods (odds ratio)
axes.vlines(0, 0, 1, colors='red', linestyles='dashed', alpha=0.5)
axes.legend(loc=(1.03, 0.712))
axes.set_xticks([-3, -2, -1, 0, 1, 2, 3])
axes.set_xticklabels([0.001, 0.01, 0.1, 1, 10, 100, 1000])
axes.set_title('Period = {0} bp'.format(kwargs['period']))
if kwargs:
xmin = kwargs['xmin']
ymin = kwargs['ymin']
xmax = kwargs['xmax']
ymax = kwargs['ymax']
axes.set_xlim(xmin, xmax)
axes.set_ylim(ymin, ymax)
plt.rcParams['savefig.facecolor'] = fig.get_facecolor()
def zoomin(cohorts, tumors):
nuceriod_plt.config_params(14)
toplot = _load(cohorts, tumors)
fig, ax = plt.subplots(nrows=2, ncols=1, figsize=(8, 7), sharex=True)
toplot.sort_values(by='qvals_snr', ascending=False, inplace=True)
for i, row in toplot.iterrows():
edgecolor = None
snr = row['snr']
if (row['qvals_snr'] < 0.05) & (snr > 8):
edgecolor = 'darkred'
if row['cross_validation_max'] < 0:
ax[1].scatter(row['peak'],
-np.log2(snr),
s=200,
linewidth=2,
edgecolor=edgecolor,
color='white',
label=row['name'],
alpha=1)
ax[1].scatter(row['peak'],
-np.log2(snr),
s=80,
edgecolor='grey',
linewidth=0.5,
color=COLORS[row['ttype']],
label=row['name'],
alpha=1)
if edgecolor == 'darkred':
ax[1].text(row['peak'] + 1, -np.log2(snr) - 0.15, row['ttype'])
for i, row in toplot.iterrows():
edgecolor = None
snr = row['snr']
if row['qvals_snr'] < 0.05 and snr > 8:
edgecolor = 'darkred'
if row['cross_validation_max'] > 0:
ax[0].scatter(row['peak'],
np.log2(snr),
s=200,
linewidth=2,
edgecolor=edgecolor,
color='white',
label=row['name'],
alpha=1)
ax[0].scatter(row['peak'],
np.log2(snr),
s=80,
edgecolor='grey',
linewidth=0.5,
color=COLORS[row['ttype']],
label=row['name'],
alpha=1)
if edgecolor == 'darkred':
ax[0].text(row['peak'] + 1, np.log2(snr) - 0.15, row['ttype'])
plt.xlabel('Period')
xlim = [i for i in range(8, 22, 2)]
ax[0].set_xticks(xlim)
ax[1].set_xticks(xlim)
yvals = [i for i in range(2, 10, 2)]
ax[0].set_yticks(yvals)
yvals = [i for i in range(-8, 0, 2)]
ax[1].set_yticks(yvals)
ylabels = [str(2**abs(i)) for i in range(2, 10, 2)]
ax[0].set_yticklabels(ylabels)
ylabels = ['{}'.format(str(2**abs(i))) for i in range(-8, 0, 2)]
ax[1].set_yticklabels(ylabels)
ax[0].spines['right'].set_visible(False)
ax[0].spines['top'].set_visible(False)
ax[1].xaxis.set_ticks_position('top')
ax[1].spines['bottom'].set_visible(False)
ax[1].spines['right'].set_visible(False)
ax[0].set_ylim(1.5, 10)
ax[1].set_ylim(-10, -1.5)
plt.tight_layout()
def zoomout(files):
nuceriod_plt.config_params(14)
df = _load(files)
df = df[df['nmuts_whole_nucleosome'] > 10000]
fig, ax = plt.subplots(nrows=2, ncols=1, figsize=(8, 7), sharex=True)
for i, row in df.iterrows():
edgecolor = None
snr = row['snr']
if row['qvals_snr'] < 0.05 and snr > 8:
edgecolor = 'darkred'
if row['cross_validation_max'] > 0:
ax[0].scatter(row['peak'],
np.log2(snr),
s=200,
linewidth=2,
edgecolor=edgecolor,
color='white',
label=row['name'],
alpha=1)
ax[0].scatter(row['peak'],
np.log2(snr),
s=80,
edgecolor='grey',
linewidth=0.5,
color=COLORS[row['name']],
label=row['name'],
alpha=1)
if edgecolor == 'darkred':
ax[0].text(row['peak'] + 10,
np.log2(snr) - 0.15, row['outname'])
for i, row in df.iterrows():
edgecolor = None
snr = row['snr']
if row['qvals_snr'] < 0.05 and snr > 8:
edgecolor = 'darkred'
if row['cross_validation_max'] < 0:
ax[1].scatter(row['peak'],
-np.log2(snr),
s=200,
linewidth=2,
edgecolor=edgecolor,
color='white',
label=row['name'],
alpha=1)
ax[1].scatter(row['peak'],
-np.log2(snr),
s=80,
edgecolor='grey',
linewidth=0.5,
color=COLORS[row['name']],
label=row['name'],
alpha=1)
if edgecolor == 'darkred':
ax[1].text(row['peak'] + 10, -np.log2(snr) - 0.15,
row['outname'])
plt.ylabel('log2(SNR)')
plt.xlabel('Period')
xlim = [i for i in range(50, 260, 20)]
ax[0].set_xticks(xlim)
ax[1].set_xticks(xlim)
yvals = [i for i in range(2, 10, 2)]
ax[0].set_yticks(yvals)
yvals = [i for i in range(-8, 0, 2)]
ax[1].set_yticks(yvals)
ylabels = [str(2**abs(i)) for i in range(2, 10, 2)]
ax[0].set_yticklabels(ylabels)
ylabels = [str(2**abs(i)) for i in range(-8, 0, 2)]
ax[1].set_yticklabels(ylabels)
ax[0].spines['right'].set_visible(False)
ax[0].spines['top'].set_visible(False)
ax[1].xaxis.set_ticks_position('top')
ax[1].spines['bottom'].set_visible(False)
ax[1].spines['right'].set_visible(False)
ax[0].set_ylim(1.5, 10)
ax[1].set_ylim(-10, -1.5)
plt.tight_layout()
def sigmoid(files, tracksheets, ttypes, colors):
nuceriod_plt.config_params(12)
df = _load(files)
toplot = df.merge(tracksheets, how='left')
toplot['ttype'] = toplot['tumor_name'].map(ttypes)
ttype_inc = collections.defaultdict(float)
total_count = 0
ttype_vals = collections.defaultdict(lambda: collections.defaultdict(list))
prop_significant = collections.defaultdict(float)
for ttype, data in toplot.groupby(by='ttype'):
if len(data) > 10:
significant = 0
d = data.sort_values(by='prop_increase_in', ascending=True)
mean_d = d['prop_increase_in'].median()
ttype_inc[ttype] = mean_d
for i, row in d.iterrows():
ttype_vals[ttype]['Prop'].append(row['prop_increase_in'])
total_count += 1
if (row['qvals_snr'] < 0.1) & (row['snr'] > 8):
significant += 1
c = 'darkred'
else:
c = 'grey'
ttype_vals[ttype]['col'].append(c)
prop_significant[ttype] = 100 * significant / len(data)
fig, ax = plt.subplots(nrows=1,
ncols=len(ttype_inc),
figsize=(17, 3.5),
sharey=True)
ax[0].set_ylabel('Relative increase in mutation rate')
ax[0].yaxis.set_ticks(np.arange(-0.3, 0.3, 0.1))
for index, t in enumerate(sorted(ttype_inc, key=ttype_inc.get)):
count = 0
for ix, val in enumerate(ttype_vals[t]['Prop']):
alpha = 0.3
if ttype_vals[t]['col'][ix] == 'darkred':
alpha = 1
ax[index].scatter(count,
val,
color=ttype_vals[t]['col'][ix],
s=10,
lw=0,
alpha=alpha)
count += 1
ax[index].hlines(ttype_inc[t],
count / 2 - count * 0.4 / 2,
count / 2 + count * 0.4 / 2,
lw=2,
color='darkgreen')
ax[index].spines['top'].set_visible(False)
ax[index].spines['right'].set_visible(False)
ax[index].spines['bottom'].set_visible(False)
ax[index].spines['left'].set_visible(False)
ax[index].set_xlabel(t, rotation=90)
ax[index].xaxis.set_ticks_position('none')
if index > 0:
ax[index].yaxis.set_ticks_position('none')
labels = [item.get_text() for item in ax[index].get_xticklabels()]
empty_string_labels = [''] * len(labels)
ax[index].set_xticklabels(empty_string_labels)
ax[index].text(
ttype_inc[t], 0.22,
'n={}\n{}%'.format(count, round(prop_significant[t], 1)))
ax[index].set_ylim(-0.2, 0.3)
ax[index].add_patch(
plt.Rectangle((0, -0.8),
count,
0.03,
color=colors[t],
lw=1,
clip_on=False,
linewidth=0))
red_dot = mlines.Line2D([], [],
color='darkred',
marker='o',
linestyle='None',
markersize=5,
label='Significant sample')
grey_dot = mlines.Line2D([], [],
color='grey',
marker='o',
linestyle='None',
markersize=5,
label='Non significant sample')
median_line = mlines.Line2D([], [],
color='darkgreen',
marker='_',
linestyle='None',
lw=40,
markersize=10,
label='Median')
plt.legend(handles=[red_dot, grey_dot, median_line],
bbox_to_anchor=[1.1, 1.1])
def plot_single(table, title):
nucperiod_plt.config_params(font_size=10)
YLIM = 60
table = pd.read_csv(table, sep='\t')
fig, ax = plt.subplots(nrows=2, ncols=4, figsize=(20, 12))
snrs = []
df = table[table['maxp'] < 19.5]
df = df[df['maxp'] > 5.5]
for i in range(6, 20):
snrs.append(df['snr_{0}p'.format(str(i))].values)
ax[0, 0].boxplot(snrs)
ax[0, 0].set_xticklabels(range(6, 20))
ax[0, 0].set_xlabel('Period (bp)')
ax[0, 0].set_ylabel('SNR')
ax[0, 0].set_title('SNR for all 1 Mb chunks')
snrs = []
counts = []
for i in range(6, 20):
df = table[table['maxp'] < i + 0.5]
df = df[df['maxp'] > i - 0.5]
counts.append(len(df))
snrs.append(df['snr_{0}p'.format(str(i))].values)
ax[1, 0].boxplot(snrs)
ax[1, 0].plot(range(1, 15), [v / 10 for v in counts], label='x10 no. chunks')
ax[1, 0].set_xticklabels(range(6, 20))
ax[1, 0].set_xlabel('Period (bp)')
ax[1, 0].set_ylabel('SNR')
ax[1, 0].set_title('SNR for 1 Mb chunks\n peaking at each period')
ax[1, 0].legend()
snrs = []
df = table[table['maxp'] < 19.5]
df = df[df['maxp'] > 5.5]
for i in range(6, 20):
snrs.append(df['fold_power_increase_{0}p'.format(str(i))].values)
ax[0, 1].boxplot(snrs)
ax[0, 1].set_xticklabels(range(6, 20))
ax[0, 1].set_ylim(-3, YLIM)
ax[0, 1].set_xlabel('Period (bp)')
ax[0, 1].set_ylabel('Fold Power Increase')
ax[0, 1].set_title('Fold power increase for all 1 Mb chunks')
snrs = []
for i in range(6, 20):
df = table[table['maxp'] < i + 0.5]
df = df[df['maxp'] > i - 0.5]
snrs.append(df['fold_power_increase_{0}p'.format(str(i))].values)
ax[1, 1].boxplot(np.array(snrs))
ax[1, 1].plot(range(1, 15), [v / 10 for v in counts], label='x10 no. chunks')
ax[1, 1].set_xticklabels(range(6, 20))
ax[1, 1].set_ylim(-3, YLIM)
ax[1, 1].set_xlabel('Period (bp)')
ax[1, 1].set_ylabel('Fold Power Increase')
ax[1, 1].set_title('Fold Power Increase for 1 Mb chunks\n peaking at each period')
ax[1, 1].legend()
snrs = []
df = table[table['maxp'] < 19.5]
df = df[df['maxp'] > 5.5]
for i in range(6, 20):
snrs.append(df['fold_snr_increase_{0}p'.format(str(i))].values)
ax[0, 2].boxplot(snrs)
ax[0, 2].set_xticklabels(range(6, 20))
ax[0, 2].set_ylim(-3, YLIM)
ax[0, 2].set_xlabel('Period (bp)')
ax[0, 2].set_ylabel('Fold SNR Increase')
ax[0, 2].set_title('Fold SNR Increase for all 1 Mb chunks')
snrs = []
for i in range(6, 20):
df = table[table['maxp'] < i + 0.5]
df = df[df['maxp'] > i - 0.5]
snrs.append(df['fold_snr_increase_{0}p'.format(str(i))].values)
ax[1, 2].boxplot(snrs)
ax[1, 2].plot(range(1, 15), [v / 10 for v in counts], label='x10 no. chunks')
ax[1, 2].set_xticklabels(range(6, 20))
ax[1, 2].set_ylim(-3, YLIM)
ax[1, 2].set_xlabel('Period (bp)')
ax[1, 2].set_ylabel('Fold SNR Increase')
ax[1, 2].set_title('Fold SNR Increase for 1 Mb chunks\n peaking at each period')
ax[1, 2].legend()
discoveries = []
for i in range(6, 20):
discoveries.append(len(table[table['logpval_power_{0}p'.format(str(i))] == 2]))
barlist = ax[0, 3].bar(list(range(6, 20)), discoveries)
barlist[list(range(6, 20)).index(10)].set_color('r')
ax[0, 3].hlines(np.mean(np.array(discoveries)), 5.6, 19.4, linestyles='dashed', colors='grey')
ax[0, 3].set_xticks(range(6, 20))
ax[0, 3].set_xlabel('Period (bp)')
ax[0, 3].set_ylabel('No. chunks significantly high in power')
ax[0, 3].set_title('Power Enrichment')
discoveries = []
for i in range(6, 20):
discoveries.append(len(table[table['logpval_snr_{0}p'.format(str(i))] == 2]))
barlist = ax[1, 3].bar(list(range(6, 20)), discoveries)
barlist[list(range(6, 20)).index(10)].set_color('r')
ax[1, 3].hlines(np.mean(np.array(discoveries)), 5.6, 19.4, linestyles='dashed', colors='grey')
ax[1, 3].set_xticks(range(6, 20))
ax[1, 3].set_ylabel('No. chunks significantly high in SNR')
ax[1, 3].set_xlabel('Period (bp)')
ax[1, 3].set_title('SNR Enrichment')
fig.suptitle(title)
plt.rcParams['savefig.facecolor'] = fig.get_facecolor()
def spectrum(original_file, simulated_files):
nucperiod_plt.config_params(font_size=14)
with open(original_file, 'rt') as f:
mydict = json.load(f)
pair_count_array = np.array(mydict['pair_count'])
motif_count = mydict['motif_count']
len_chunk = mydict['chunk_len']
# define the signal
signal = np.array([(v / len_chunk) / (motif_count / len_chunk) ** 2 for v in pair_count_array])
# define figure
figsize = (10, 5)
fig_spec, ax_spec = plt.subplots(1, 1, figsize=figsize)
ax_spec.set_xlabel('period (bp)')
ax_spec.set_ylabel('power')
# 3-smoothing
signal = spectral.mean_smooth(signal, 3)
counter = 0
# detrended smooth autocorrelation
initial_values_dict = {'a_0': np.mean(signal[4:]), 'a_1': 0., 'a_2': 0.}
params, obj_func = non_linear.create_quadratic_model(initial_values_dict)
x = np.arange(len(signal[4:]))
non_linear_fitter = non_linear.NonLinearFitting(obj_func, params, x, signal[4:])
_, predicted = non_linear_fitter.least_squares()
signal_detrended = signal[4:] - predicted
for file in simulated_files:
counter += 1
with open(file, 'rt') as f:
random_chunk = json.load(f)
pc = np.array(random_chunk['pair_count'])
mc = random_chunk['motif_count']
len_random_chunk = random_chunk['chunk_len']
random_signal = np.array([(v / len_chunk) / (mc / len_random_chunk) ** 2 for v in pc])
random_signal = spectral.mean_smooth(random_signal, 3)
# detrended smooth autocorrelation
initial_values_dict = {'a_0': np.mean(random_signal[4:]), 'a_1': 0., 'a_2': 0.}
params, obj_func = non_linear.create_quadratic_model(initial_values_dict)
x = np.arange(len(random_signal[4:]))
non_linear_fitter = non_linear.NonLinearFitting(obj_func, params, x, random_signal[4:])
_, predicted = non_linear_fitter.least_squares()
random_signal_detrended = random_signal[4:] - predicted
# DTFT
if counter == 1:
label = 'randomized'
else:
label = None
dtft_spectrum_plot(random_signal_detrended, ax_spec, title='periodogram',
norm=False, color='grey', alpha=0.5, label=label)
dtft_spectrum_plot(signal_detrended, ax_spec, title='periodogram', norm=False, label='observed')
ax_spec.legend()
plt.rcParams['savefig.facecolor'] = fig_spec.get_facecolor()