Skip to content
This repository has been archived by the owner on Jun 7, 2023. It is now read-only.

Commit

Permalink
Merge pull request #73 from tdincer/main
Browse files Browse the repository at this point in the history 
blackify
  • Loading branch information
@kabilar
kabilar committed Jun 3, 2022
2 parents ac873d9 + c0a56cb commit 5c75919
Show file tree
Hide file tree
Showing 7 changed files with 251 additions and 153 deletions.
1 change: 0 additions & 1 deletion workflow_calcium_imaging/__init__.py
View file
Original file line number Diff line number Diff line change
@@ -1 +0,0 @@

97 changes: 57 additions & 40 deletions workflow_calcium_imaging/analysis.py
View file
Original file line number Diff line number Diff line change
@@ -1,11 +1,16 @@
import datajoint as dj
import numpy as np
from workflow_calcium_imaging.pipeline import (
db_prefix,
session,
scan,
imaging,
trial,
event,
)

from workflow_calcium_imaging.pipeline import db_prefix, session, scan, imaging, trial, \
event


schema = dj.schema(db_prefix + 'analysis')
schema = dj.schema(db_prefix + "analysis")


@schema
Expand Down Expand Up @@ -44,51 +49,57 @@ class AlignedTrialActivity(dj.Part):
"""

def make(self, key):
sess_time, scan_time, nframes, frame_rate = (scan.ScanInfo * session.Session
& key
).fetch1('session_datetime',
'scan_datetime',
'nframes', 'fps')
sess_time, scan_time, nframes, frame_rate = (
scan.ScanInfo * session.Session & key
).fetch1("session_datetime", "scan_datetime", "nframes", "fps")

# Estimation of frame timestamps with respect to the session-start
# (to be replaced by timestamps retrieved from some synchronization routine)
scan_start = (scan_time - sess_time).total_seconds() if scan_time else 0
frame_timestamps = np.arange(nframes) / frame_rate + scan_start

trialized_event_times = trial.get_trialized_alignment_event_times(
key, trial.Trial & (ActivityAlignmentCondition.Trial & key))
key, trial.Trial & (ActivityAlignmentCondition.Trial & key)
)

min_limit = (trialized_event_times.event - trialized_event_times.start).max()
max_limit = (trialized_event_times.end - trialized_event_times.event).max()

aligned_timestamps = np.arange(-min_limit, max_limit, 1/frame_rate)
aligned_timestamps = np.arange(-min_limit, max_limit, 1 / frame_rate)
nsamples = len(aligned_timestamps)

trace_keys, activity_traces = (imaging.Activity.Trace & key
).fetch('KEY', 'activity_trace', order_by='mask')
trace_keys, activity_traces = (imaging.Activity.Trace & key).fetch(
"KEY", "activity_trace", order_by="mask"
)
activity_traces = np.vstack(activity_traces)

aligned_trial_activities = []
for _, r in trialized_event_times.iterrows():
if r.event is None or np.isnan(r.event):
continue
alignment_start_idx = int((r.event - min_limit) * frame_rate)
roi_aligned_activities = activity_traces[:,
alignment_start_idx:
(alignment_start_idx + nsamples)]
roi_aligned_activities = activity_traces[
:, alignment_start_idx : (alignment_start_idx + nsamples)
]
if roi_aligned_activities.shape[-1] != nsamples:
shape_diff = nsamples - roi_aligned_activities.shape[-1]
roi_aligned_activities = np.pad(roi_aligned_activities,
((0, 0), (0, shape_diff)),
mode='constant', constant_values=np.nan)

aligned_trial_activities.extend([{**key, **r.trial_key, **trace_key,
'aligned_trace': aligned_trace}
for trace_key, aligned_trace
in zip(trace_keys,
roi_aligned_activities)])

self.insert1({**key, 'aligned_timestamps': aligned_timestamps})
roi_aligned_activities = np.pad(
roi_aligned_activities,
((0, 0), (0, shape_diff)),
mode="constant",
constant_values=np.nan,
)

aligned_trial_activities.extend(
[
{**key, **r.trial_key, **trace_key, "aligned_trace": aligned_trace}
for trace_key, aligned_trace in zip(
trace_keys, roi_aligned_activities
)
]
)

self.insert1({**key, "aligned_timestamps": aligned_timestamps})
self.AlignedTrialActivity.insert(aligned_trial_activities)

def plot_aligned_activities(self, key, roi, axs=None, title=None):
Expand All @@ -109,25 +120,31 @@ def plot_aligned_activities(self, key, roi, axs=None, title=None):
else:
ax0, ax1 = axs

aligned_timestamps = (self & key).fetch1('aligned_timestamps')
trial_ids, aligned_spikes = (self.AlignedTrialActivity
& key & {'mask': roi}).fetch(
'trial_id', 'aligned_trace', order_by='trial_id')
aligned_timestamps = (self & key).fetch1("aligned_timestamps")
trial_ids, aligned_spikes = (
self.AlignedTrialActivity & key & {"mask": roi}
).fetch("trial_id", "aligned_trace", order_by="trial_id")

aligned_spikes = np.vstack(aligned_spikes)

ax0.imshow(aligned_spikes, cmap='inferno',
interpolation='nearest', aspect='auto',
extent=(aligned_timestamps[0],
aligned_timestamps[-1],
0,
aligned_spikes.shape[0]))
ax0.axvline(x=0, linestyle='--', color='white')
ax0.imshow(
aligned_spikes,
cmap="inferno",
interpolation="nearest",
aspect="auto",
extent=(
aligned_timestamps[0],
aligned_timestamps[-1],
0,
aligned_spikes.shape[0],
),
)
ax0.axvline(x=0, linestyle="--", color="white")
ax0.set_axis_off()

ax1.plot(aligned_timestamps, np.nanmean(aligned_spikes, axis=0))
ax1.axvline(x=0, linestyle='--', color='black')
ax1.set_xlabel('Time (s)')
ax1.axvline(x=0, linestyle="--", color="black")
ax1.set_xlabel("Time (s)")
ax1.set_xlim(aligned_timestamps[0], aligned_timestamps[-1])

if title:
Expand Down
Loading

0 comments on commit 5c75919

Please sign in to comment.