Skip to content

seung-lab/e2198_Ca_imaging

Repository files navigation

e2198_Ca_imaging

This repository contains code and data for two analyses:

  • Analysis of calcium imaging data and finding the cell correspondance between the calcium imaging and the EM reconstruction.
  • OBSOLETE - Cell clustering based on anatomy from EM reconstruction and physiology from calcium imaging.

Data

EM vs Calcium

  • code/cell_mapping_verified.m
    • The cell_dict_j variable in this file contains the correspondance between EM cell IDs and calciums ROI indices checked by a human expert.

Data for clustering and other anatomical data

  • data/cell_info_clustering.mat - OBSOLETE

    • All derived data needed for the clustering task. Last update: 2016/4/1
  • data/soma_coords_warped_20160401cleanup.mat

    • Soma center coordinates transformed to the warped coordinate system.
  • data/completed_051016_measurements.csv

    • Soma radius and corrected segmentation volume (omnivol) for each cellid Currently contains info for all 396 cells

Ca imaging data

Derived data
  • data/ca_dsos.20160822.mat

    • Direction selectivity and orientation selectivity computed from the parameter fits of the calcium traces.
  • data/coeffs16.20160822.mat

    • Parameter fits to the calcium traces of each ROI. Specifically, coeffs16{3,2} is the double exponential fit with tau, amplitudes and crossing times as parameters. #TODO: details
Calcium trace data
  • data/roi_data.mat

    • Information pertaining to the calcium imaging in easy to access formats, including calcium traces time-aligned and grouped into individual ROIs (100*DeltaF/F, before baseline detrending).
    • Running the reorganize.m script after loading this file will generate some additional variables and statistics on these data.
  • data/roi_data_stimconsts.mat

    • Some addition information regarding the stimulus in the calcium imaging that are used to generate data in roi_data.mat and needed for making summary plots.
Raw data
  • data/0**_p2_g7_DSbars_200um_export.mat

    • Calcium signals processed into ROIs but not clipped (not time-aligned).
  • data/sDS 8x45deg, narrow, 4.0s.QDS

    • Stimulus description script.
  • data/AVG_043_p2_g7_ZoomedOut_center.tif

    • overview image
  • data/AVG_4classes_043_p2_g7_ZoomedOut_center_011.rpb

    • ROIs
  • rawdata/

    • These are the raw calcium imaging data.

Clustering

Ca imaging

Ca imaging data from Briggman et al. (2011).
The stimulus is a bright bar on a dark background. "Each of the eight stimuli used was a bar (0.2 mm wide x 1 mm long) moving in one of eight different directions at 1 mm/sec." So the On and Off responses to the bar edges should be 1 second apart.

128 ms/frame, 31 frames per stimulus

data/roi_data.mat contains DeltaF/F for all ROIs, and some metadata

"The imaged field of view for each stimulus presentation was 100 μm × 100 μm. We acquired data from nine such fields, arranged in a 3 × 3 mosaic, for a total recorded area of 300 μm × 300 μm."

In the "data/sDS 8x45deg, narrow, 4.0s.QDS" Note the 6th condition (direction) has a duration of 3.5s instead of 3.95s.

In the overview image (with the blood vessel ‘V’ pattern opening to the right): 0 deg (stimulus label) is left (yellow cells in Briggman et al. 2011), 90 deg is up (red cells) 180 is right (green cells) 270 is down (magenta cells)

Polar Coordinates

Final unified coordinates for paper:
  • 0 ~ rostral
  • 270 ~ ventral
Coordinates in code and data:

Stimuli coords (as labeled in "data/sDS 8x45deg, narrow, 4.0s.QDS"):

  • 180 ~ ventral (2an Ca ~ 170deg)
  • 270 ~ rostral (7o Ca ~ 250deg)

EM (Omni) y-z coord:

  • 0 = +y ~ rostral
  • 90 = +z ~ ventral

About

Cell clustering and calcium imaging analysis of e2198

Resources

Stars

Watchers

Forks

Packages

No packages published