Calcium imaging during behavioural sessions in which mice are required to detect two-photon optogenetic activation of varying numbers of excitatory neurons in L2/3 barrel cortex.

SummaryRaw extracted two-photon calcium imaging and two-photon photostimulation data from: Dalgleish HWP*, Russell LE*, Packer AM*, Roth A, Gauld OM, Greenstreet F, Thompson EJ, Hausser M (2020). How many neurons are sufficient for perception of cortical activity? eLife If you use these data in a paper, please cite the above Dalgleish et al. 2020 (eLife) paper and reference this Figshare dataset DOI:https://doi.org/10.6084/m9.figshare.13128950 Data for import, analysis and figure plotting can be found here:https://github.com/alloptical/Dalgleish-eLife-2020Please follow instructions in the Raw data download and import section to unzip correctly for use with this repository. Please contact hwpdalgleish@gmail.com for any questions. DescriptionThis dataset contains the raw extracted two-photon calcium imaging data (Suite2p – ROIs, traces, metadata), photostimulation protocols, opsin expression data and synchronisation data from “number of neurons” psychometric curve behavioural sessions. During these sessions animals were required to detect two-photon optogenetic activation of varying numbers of cortical excitatory neurons by licking for sucrose rewards at a water spout monitored by an electrical lick sensor. Each zip file is data from 1 animal, 1 behavioural session on a given day (~1 – 1.5 hours) with the naming convention sessionDate_animalID. Experiments are in L2/3 barrel cortex of mice head-fixed on a cylindrical treadmill beneath a two-photon microscope. Imaging is of GCaMP6s at 920 nm (~50 mW) across 4 planes (using an ETL) at ~27 Hz frame rate, ~7 Hz volume rate. Photostimulation is of C1V1-Kv2.1 (somatically restricted) at 1030 nm (6 mW/neuron). All animal IDs beginning “L” are Emx1-Cre;CaMKIIa-tTA;Ai94 GCaMP6s transgenic mice injected with AAV2/9-CaMKII-C1V1(t/t)-mRuby2-Kv2.1. All animal IDs beginning “OG” are wild-type (C57/BL6) mice injected with AAV1-Syn-GCaMP6s-WPRE-SV40 and AAV2/9-CaMKII-C1V1(t/t)-mScarlett-Kv2.1. FormatFall.mat – Suite2p output (ROIs, traces etc.) in standard format for Python Suite2p’s MATLAB output (see Suite2p documentation for details). ROIs have been manually curated (NB to use curated ROIs use the iscell label). targets – photostimulation target data. Each *_Points.mat file returns a points variable with fields X, Y, Z and Zum corresponding to XY co-ordinate in pixels, Z co-ordinate in plane number and Z co-ordinate in µm respectively. Each file corresponds to a different stimulus type (number of neurons targeted) used during the experiment (see number of elements in the above fields). The *_VarFile_*.mat file contains the training photostimulation protocol. This can be synced with the *_Points.mat files (see above) and *_paq_analysis.mat file (see below) via code in the Dalgleish et al. 2020 Github repo. cellPose – C1V1 expression images and cellpose-identified C1V1-expressing neurons. *.tif files are acquired expression images of C1V1-mCherry (765nm). *_cellPoseCentroids.mat files contain co-ordinates of C1V1-expressing neurons (NB this has a similar format to the target *_Points.mat files described above). Other files are raw output from the Cellpose algorithm (see Cellpose documentation for details). *_paqanalysis.mat – synchronisation data recording the timing of all features of the experiment. Fields should be self-explanatory. Returns variable pa with relevant fields: frames: imaging framesstims: stimulus times as a structure array where each instance records timing for a given stimulus channel (see NB below). Two trigger types: “in” are sent from the behavioural control hardware (i.e. “deliver photostimulation”); “out” are sent from the microscope photostimulation hardware (i.e. “photostimulation delivered”). Where possible use “out” for the most accurate timing information (e.g. for Go stimulus trials). NB Catch trials only have “in” triggers as no photostimulation was delivered.licks: lick timesrewards: reward delivery timesrunning: rotary encoder reading from linear treadmill NB that there are two “stimulus types” in our experiment and thus two relevant stimulus channels: Go trials, with photostimulation, delivered via channel 7 referenced via pa.stims(7), and Catch trials, with no stimulus, delivered via channel 6 referenced via pa.stims(6). Go trials have 7 different trial types, or variations/var, corresponding to different numbers of neurons stimulated. These are:1: 200 neurons2: 100 neurons3: 75 neurons4: 50 neurons5: 25 neurons6: 10 neurons7: 5 neurons