Matthew A. Churgin, Danylo Lavrentovich

de Bivort Lab

Experimental data and analysis scripts for A neural correlate of individual odor preference in Drosophila.

This GitHub repository contains raw behavioral and 2-photon calcium data, along with MATLAB scripts that generate the figures. Mappings between MATLAB scripts and figure panels are listed at the above link. A complete repository, with large raw microscopy images too large for sharing via GitHub, is available by contacting the authors.

A brief walkthrough of the directories in this repository:

utilities/ contains helpful auxiliary functions and .mat files used throughout the repository. If you download this repository, please set your local path in SET_DIRECTORY_PATH.m and run it to generate analysis_dir_path.mat, which is loaded by all MATLAB analysis scripts to keep file paths consistent

ORNvsPN_analysis_ALLDATA/, ORN_analysis_oct_vs_air_choice/, ORN_analysis/, PN_analysis_oct_vs_air_choice/, PN_analysis/ store calcium imaging data and perform analyses. Main analysis script that generates most of the figures is ORNvsPN_analysis_ALLDATA/plotRandomSubsetPNandORN.m

IHC/ contains Brp-Short immunohistochemistry analysis

r2_latent_inference/ contains Python scripts that run simulations to infer the extent of latent neural predictor - latent behavior correlation based on the calcium/Brp-Short linear models