def plot_behavior_events_trace(session,
xmin=360,
length=3,
ax=None,
save_dir=None):
xmax = xmin + 60 * length
interval = 20
if ax is None:
figsize = (15, 4)
fig, ax = plt.subplots(figsize=figsize)
ax.plot(session.running_speed.timestamps,
session.running_speed.values,
color=sns.color_palette()[0])
ax = add_stim_color_span(session, ax, xlim=[xmin, xmax])
ax = plot_behavior_events(session, ax)
ax = restrict_axes(xmin, xmax, interval, ax)
ax.set_ylabel('running speed (cm/s)')
ax.set_xlabel('time (sec)')
if save_dir:
fig.tight_layout()
ut.save_figure(
fig, figsize, save_dir, 'behavior_events',
str(session.metadata['ophys_experiment_id']) + '_' + str(xmin))
plt.close()
return ax
def plot_max_proj_and_roi_masks(session, save_dir=None):
figsize = (15, 5)
fig, ax = plt.subplots(1, 3, figsize=figsize)
ax = ax.ravel()
ax[0].imshow(session.max_projection,
cmap='gray',
vmin=0,
vmax=np.amax(session.max_projection))
ax[0].axis('off')
ax[0].set_title('max intensity projection')
ax[1].imshow(session.segmentation_mask_image, cmap='gray')
ax[1].set_title('roi masks')
ax[1].axis('off')
ax[2].imshow(session.max_projection,
cmap='gray',
vmin=0,
vmax=np.amax(session.max_projection))
ax[2].axis('off')
ax[2].set_title(str(session.metadata['ophys_experiment_id']))
tmp = session.segmentation_mask_image.data.copy()
mask = np.empty(session.segmentation_mask_image.data.shape, dtype=np.float)
mask[:] = np.nan
mask[tmp > 0] = 1
cax = ax[2].imshow(mask, cmap='hsv', alpha=0.4, vmin=0, vmax=1)
if save_dir:
ut.save_figure(fig, figsize, save_dir, 'roi_masks',
str(session.metadata['ophys_experiment_id']))
def plot_mean_trace_heatmap(mean_df,
ax=None,
save_dir=None,
window=[-4, 8],
interval_sec=2):
"""
There must be only one row per cell in the input df.
For example, if it is a mean of the trial_response_df, select only trials where go=True before passing to this function.
"""
data = mean_df[mean_df.pref_stim == True].copy()
if ax is None:
figsize = (3, 6)
fig, ax = plt.subplots(1, 1, figsize=figsize)
order = np.argsort(data.mean_response.values)[::-1]
cells = data.cell_specimen_id.unique()[order]
len_trace = len(data.mean_trace.values[0])
response_array = np.empty((len(cells), len_trace))
for x, cell_specimen_id in enumerate(cells):
tmp = data[data.cell_specimen_id == cell_specimen_id]
if len(tmp) >= 1:
trace = tmp.mean_trace.values[0]
else:
trace = np.empty((len_trace))
trace[:] = np.nan
response_array[x, :] = trace
sns.heatmap(data=response_array,
vmin=0,
vmax=np.percentile(response_array, 99),
ax=ax,
cmap='magma',
cbar=False)
xticks, xticklabels = ut.get_xticks_xticklabels(trace,
31.,
interval_sec=interval_sec,
window=window)
ax.set_xticks(xticks)
ax.set_xticklabels([int(x) for x in xticklabels])
if response_array.shape[0] < 50:
interval = 10
else:
interval = 50
ax.set_yticks(np.arange(0, response_array.shape[0], interval))
ax.set_yticklabels(np.arange(0, response_array.shape[0], interval))
ax.set_xlabel('time after change (s)', fontsize=16)
ax.set_ylabel('cells')
if save_dir:
fig.tight_layout()
ut.save_figure(fig, figsize, save_dir, 'experiment_summary',
'mean_trace_heatmap_' + condition + suffix)
return ax
def plot_experiment_summary_figure(session, save_dir=None):
import allensdk.brain_observatory.behavior.swdb.utilities as ut
meta = session.metadata
title = meta['driver_line'][0] + ', ' + meta['targeted_structure'] + ', ' + str(meta['imaging_depth']) + ', ' + \
session.task_parameters['stage']
interval_seconds = 600
ophys_frame_rate = int(session.metadata['ophys_frame_rate'])
figsize = [2 * 11, 2 * 8.5]
fig = plt.figure(figsize=figsize, facecolor='white')
ax = placeAxesOnGrid(fig, dim=(1, 1), xspan=(.0, .2), yspan=(0, .2))
ax.imshow(session.max_projection, cmap='gray')
ax.set_title('max intensity projection')
ax.axis('off')
ax = placeAxesOnGrid(fig, dim=(1, 1), xspan=(0, .18), yspan=(.24, .4))
trials = session.trials.copy()
trials = trials[trials.reward_rate > 1]
plot_transitions_response_heatmap(trials, ax=ax)
ax = placeAxesOnGrid(fig, dim=(1, 1), xspan=(.24, .86), yspan=(0, .26))
ax = plot_traces_heatmap(session, ax=ax)
ax.set_title(title)
ax = placeAxesOnGrid(fig, dim=(1, 1), xspan=(.28, .92), yspan=(.32, .44))
ax.plot(session.running_speed.timestamps, session.running_speed.values)
ax.set_xlabel('time (seconds)')
ax.set_ylabel('running speed\n(cm/s)')
ax = placeAxesOnGrid(fig, dim=(1, 1), xspan=(.86, 1.), yspan=(0, .2))
image_index = 0
ax.imshow(session.stimulus_templates[image_index, :, :], cmap='gray')
st = session.stimulus_presentations.copy()
image_name = st[st.image_index == image_index].image_name.values[0]
ax.set_title(image_name)
ax.axis('off')
ax = placeAxesOnGrid(fig, dim=(1, 1), xspan=(.0, .17), yspan=(.54, .99))
ax = plot_lick_raster(session.trials, ax=ax)
ax = placeAxesOnGrid(fig, dim=(1, 1), xspan=(.24, .42), yspan=(.54, .99))
fr = session.flash_response_df
mdf = ut.get_mean_df(fr, conditions=['cell_specimen_id', 'image_name'])
plot_mean_image_response_heatmap(mdf, title=None, ax=ax)
ax = placeAxesOnGrid(fig, dim=(1, 1), xspan=(.52, .68), yspan=(.54, .99))
tr = session.trial_response_df.copy()
mdf = ut.get_mean_df(tr[tr.go], conditions=['cell_specimen_id'])
mdf['pref_stim'] = True
ax = plot_mean_trace_heatmap(mdf, ax=ax, window=[-4, 8], interval_sec=2)
ax.set_title('mean trace for pref image')
ax.set_ylabel('cells')
ax = placeAxesOnGrid(fig, dim=(1, 1), xspan=(.76, .98), yspan=(.5, .62))
ax.plot(session.trials.reward_rate)
ax.set_ylabel('reward rate')
ax.set_xlabel('trials')
ax = placeAxesOnGrid(fig, dim=(1, 1), xspan=(.76, 0.98), yspan=(.68, .8))
plot_behavior_segment(session, xlims=[620, 640], ax=ax)
ax = placeAxesOnGrid(fig, dim=(1, 1), xspan=(.76, .98), yspan=(.86, .99))
traces = tr[(tr.go == True)].dff_trace.values
ax = ut.plot_mean_trace(traces, window=[-4, 8], ax=ax)
ax = ut.plot_flashes_on_trace(ax, window=[-4, 8], go_trials_only=True)
ax.set_xlabel('time after change (sec)')
ax.set_ylabel('mean dF/F')
fig.tight_layout()
if save_dir:
fig.tight_layout()
ut.save_figure(fig, figsize, save_dir, 'experiment_summary',
str(experiment_id))
def plot_mean_image_response_heatmap(mean_df,
title=None,
ax=None,
save_dir=None):
df = mean_df.copy()
if 'change_image_name' in df.keys():
image_key = 'change_image_name'
else:
image_key = 'image_name'
images = np.sort(df[image_key].unique())
cell_list = []
for image in images:
tmp = df[(df[image_key] == image) & (df.pref_stim == True)]
order = np.argsort(tmp.mean_response.values)[::-1]
cell_ids = list(tmp.cell_specimen_id.values[order])
cell_list = cell_list + cell_ids
response_matrix = np.empty((len(cell_list), len(images)))
for i, cell in enumerate(cell_list):
responses = []
for image in images:
response = df[(df.cell_specimen_id == cell)
& (df[image_key] == image)].mean_response.values[0]
responses.append(response)
response_matrix[i, :] = np.asarray(responses)
if ax is None:
figsize = (4, 7)
fig, ax = plt.subplots(figsize=figsize)
vmax = 0.3
label = 'mean dF/F'
ax = sns.heatmap(response_matrix,
cmap='magma',
linewidths=0,
linecolor='white',
square=False,
vmin=0,
vmax=vmax,
robust=True,
cbar_kws={
"drawedges": False,
"shrink": 1,
"label": label
},
ax=ax)
if title is None:
title = 'mean response by image'
ax.set_title(title, va='bottom', ha='center')
ax.set_xticklabels(images, rotation=90)
ax.set_ylabel('cells')
if response_matrix.shape[0] < 50:
interval = 10
else:
interval = 50
ax.set_yticks(np.arange(0, response_matrix.shape[0], interval))
ax.set_yticklabels(np.arange(0, response_matrix.shape[0], interval))
if save_dir:
fig.tight_layout()
ut.save_figure(fig, figsize, save_dir, 'experiment_summary',
'mean_image_response_heatmap' + suffix)
def plot_example_traces_and_behavior(session,
xmin_seconds,
length_mins,
cell_label=False,
save_dir=None):
traces = np.stack(session.dff_traces.dff.values)
cell_indices = ut.get_active_cell_indices(traces)
interval_seconds = 10
xmax_seconds = xmin_seconds + (length_mins * 60) + 1
xlim = [xmin_seconds, xmax_seconds]
figsize = (14, 10)
fig, ax = plt.subplots(len(cell_indices) + 1, 1, figsize=figsize)
ax = ax.ravel()
ymins = []
ymaxs = []
for i, cell_index in enumerate(cell_indices):
ax[i].tick_params(reset=True,
which='both',
bottom='off',
top='off',
right='off',
left='off',
labeltop='off',
labelright='off',
labelleft='off',
labelbottom='off')
ax[i] = plot_trace(session.ophys_timestamps,
traces[cell_index, :],
ax=ax[i],
title='',
ylabel=str(cell_index),
color=[.5, .5, .5])
ax[i] = add_stim_color_span(session, ax=ax[i], xlim=xlim)
ax[i] = restrict_axes(xmin_seconds,
xmax_seconds,
interval_seconds,
ax=ax[i])
ax[i].set_xticks([])
ax[i].set_xlabel('')
ax[i].set_xlim(xlim)
ymin, ymax = ax[i].get_ylim()
ymins.append(ymin)
ymaxs.append(ymax)
if cell_label:
ax[i].set_ylabel('cell ' + str(i), fontsize=12)
else:
ax[i].set_ylabel('')
ax[i].set_yticks([])
sns.despine(ax=ax[i], left=True, bottom=True)
ymin, ymax = ax[i].get_ylim()
if 'Vip' in session.metadata['full_genotype']:
ax[i].vlines(x=xmin_seconds, ymin=0, ymax=2, linewidth=4)
ax[i].set_ylim(ymin=-0.5, ymax=5)
elif 'Slc' in session.metadata['full_genotype']:
ax[i].vlines(x=xmin_seconds, ymin=0, ymax=1, linewidth=4)
ax[i].set_ylim(ymin=-0.5, ymax=3)
ax[i].get_xaxis().set_ticks([])
ax[i].get_yaxis().set_ticks([])
ax[i].tick_params(which='both',
bottom='off',
top='off',
right='off',
left='off',
labeltop='off',
labelright='off',
labelleft='off',
labelbottom='off')
ax[i].set_xticklabels('')
i += 1
ax[i].tick_params(axis="x",
bottom=True,
top=False,
labelbottom=True,
labeltop=False)
ax[i].plot(session.running_speed.timestamps,
session.running_speed.values,
color=sns.color_palette()[0])
ax[i] = plot_behavior_events(session, ax=ax[i])
ax[i] = add_stim_color_span(session, ax=ax[i], xlim=xlim)
ax[i] = restrict_axes(xmin_seconds,
xmax_seconds,
interval_seconds,
ax=ax[i])
ax[i].set_xlim(xlim)
ax[i].set_ylabel('run speed\n(cm/s)', fontsize=12)
sns.despine(ax=ax[i], left=True, bottom=True)
ax[i].set_yticklabels('')
xticks = np.arange(xmin_seconds, xmax_seconds, interval_seconds)
ax[i].set_xticks(xticks)
ax[i].set_xticklabels(xticks)
ax[i].set_xlabel('time (seconds)')
ax[0].set_title(
str(session.metadata['ophys_experiment_id']) + '_' +
session.metadata['full_genotype'].split('-')[0])
plt.subplots_adjust(wspace=0, hspace=0)
plt.subplots_adjust(bottom=0.2)
if save_dir:
ut.save_figure(
fig, figsize, save_dir, 'example_traces',
str(session.metadata['ophys_experiment_id']) + '_' + str(xlim[0]))