Companion Data Page
Idiosyncratic neural coding and neuromodulation of olfactory individuality in Drosophila
Kyle Honneger1,2*, Matthew Smith1*, Matthew Churgin1, Glenn Turner3, Benjamin de Bivort1
1 - Department of Organismic and Evolutionary Biology and Center for Brain Science, Harvard University, Cambridge, Massachusetts, USA.
2 - Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, Illinois, USA.
3 - Janelia Research Campus, Howard Hughes Medical Institute, Ashburn, VA, USA.
* - contributed equally.
Manuscript: hosted at Proceedings of the National Academy of Sciences
Repository of data and scripts (Zenodo)
|Zenodo repository||Archive .zip file containing all experimental data and analysis scripts.|
|controlCalciumResponses.mat||Archive .mat file containing a cell-array titled "flyTimeCourse". Each response consists of 20 time points of dF/F Ca2+ response to a single odor for a fly in control conditions. Inside each fly's cell there are 15 rows, each row represents a specific glomerulus. If that glomerulus was not identified in that fly then the response data will contain all "NaNs". There are 2 columns, column 1 represents the response to odor panel trial-1, and column 2 represents the response to odor panel trial-2.|
|amwCalciumResponses.mat||Same as above, but for flies fed food supplemented with amw.|
|controlVolumes.zip||Archive .mat file of identified glomeruli for each individual fly. An identified glomerulus is stored as a 3D binary array. The variable "controlVar" contains a cell-arrays for each fly. Within each fly's cell-array there are glomeruli identified during odor panel trial-1 and odor panel trial-2, column 1 and column 2 respectively. Within each trial there are 3D binary arrays for the 15 potentially identified glomeruli for each fly. If a specific glomeruli is not identified for that trial then the 3D array will contain all zeros.|
|amwVolumes.zip||Same as above, but for flies fed food supplemented with amw.|
Odor preference data
|24h-persist.mat||A .mat file that contains 24 hour persistence data from 141 flies. The file contains the variable "daypersist", which is a 141x2 array. Each row represents a unique fly. Column 1 represents odor preference for OCT on day one, and column 2 represents odor preference for the same fly on day two.|
|nullOdorPreferenceDatasets.zip||A .zip containing .mat files for datasets of flies' behavior in the odor tunnels with either air vs air conditions, or air vs odor conditions. The isogenic control fly line, isoKH11*, is used in both datasets.|
|flat_files.zip||A .zip containing .csv files for behavioral experiments. The title of the .csv file details the experimental manipulation, temperature, and thermo-genetics used (if any). The values within the file are odor preference scores for MCH. Each row represents a unique fly's odor preference score.|
|Stan modeling data.zip||A .zip containing .rds files for Bayesan linear modeling of behavior in Stan (statistical modeling software)|
|stan_models.zip||A .zip containing .rds files for outputs of Bayesian linear models|
Imaging analysis and figure generation
|qwkGlomeruliViewer.m||A MATLAB function to view the identified glomeruli for each fly. The function takes two inputs: volumes, FlyIDs. It uses the input FlyID to index into the volumes cell-array and generate figures of the glomeruli identified for each fly across odor panel trials 1 and 2. The first input, "volumes", is the cell-array in the controlVolumes.zip or amwVolumes.zip. The second input "FlyIDs" is an array of numbers corresponding to the flies whose glomeruli will be displayed. If this input is left empty, by default the function will generate figures for all flies in the volumes cell-array.|
|individualityNeuralCoding_control.zip||Archive .zip file containing scripts and data to perform PCA on the responses to 12 odors across flies grown in control conditions and generate the scatter plot of odor representations in PC-space. The script will also calculate distances between repeated trials within an individual and distances between individuals. The error bars on these estimates are bootstrapped with 10,000 iterations. The within individual variability (intra-fly) and between individual variability (inter-fly) distances are displayed as bar graphs. To access, unzip the file, index into the folder, and run the run.m file.|
|individualityNeuralCoding_amw.zip||Same as above but for flies treated with amw|
|individualityNeuralCoding_control_OCTMCH.zip||Archive .zip file containing scripts and data to perform PCA on the responses specifically to octanol and MCH across flies grown in control conditions and generate the scatter plot of odor representations in PC-space. The script will also calculate distances between repeated trials within an individual and distances between individuals. The error bars on these estimates are bootstrapped with 10,000 iterations. The within individual variability (intra-fly) and between individual variability (inter-fly) distances are displayed as bar graphs. To access, unzip the file, index into the folder and run the run.m file.|
|individualityNeuralCoding_amw_OCTMCH.zip||Same as above but for flies treated with amw|
Bayesian modeling scripts and data
|behaviorScripts.zip||.m, .R, and .stan files for data analysis and figure generation for the odor behavior.|