# DrosophilaOlfaction

A repository containinng the complete set of data to reproduce the results presented in the manuscript titled **_Olfactory responses of Drosophila are encoded in the organization of projection neurons_**

Copyright 2022 Kiri Choi

## Description

The scripts utilize two datasets: the FAFB dataset by [Bates et al. 2020](https://www.sciencedirect.com/science/article/pii/S0960982220308587) and the hemibrain dataset by [Scheffer et al. 2020](https://elifesciences.org/articles/57443).
A part of these datasets is reproduced in the repository (https://github.com/kirichoi/DrosophilaOlfaction), which are necessary to run these scripts. 
For the datasets, we would like to credit all the original authors.
Below is a short description of what each file contains:

- *Drosophila_FAFB.py*: the Python script for the data analysis based on uPNs that innervate all three neuropils in the FAFB dataset
- *Drosophila_FAFB_other_uPN.py*: the Python script for the data analysis based on uPNs that do not innervate all three neuropils in the FAFB dataset
- *Drosophila_labeled_line.py*: the Python script for the labeled line study
- *Drosophila_labeled_line_high_thres.py*: the Python script for the labeled line study using higher synaptic weight threshold
- *Drosophila_neuprint.py*: the Python script for the data analysis based on uPNs that innervate all three neuropils in the hemibrain dataset
- *import_neuprint.py*: the Python script querying the neurons used in the study from the neuPrint database
- *FAFB_summary.xlsx*: a summary of the FAFB dataset analysis result
- *hemibrain_summary.xlsx*: a summary of the hemibrain dataset analysis result
- *FAFB*
    - *FAFB_swc*: contains the neuron reconstructions in .swc format from the FAFB dataset
    - *morph_dist_(AL/MB/LH)_r_FAFB(.csv/.npy)*: precomputed distance matrics of uPNs innervating all three neuropils
    - *morph_dist_(AL/MB/LH)_r_FAFB_uPN(.csv/.npy)*: precomputed distance matrics of all uPNs innervating each neuropil
    - *NBLAST_(AL/MB/LH)_FAFB(.csv/.npy)*: precomputed NBLAST distance matrics of uPNs innervating all three neuropils
- *hemibrain*
    - *neuprint_PN_invte_all*: contains the neuron reconstructions in .swc format from the hemibrain dataset
    - *morph_dist_(AL/MB/LH)_r_neuprint(.csv/.npy)*: precomputed distance matrics of uPNs innervating all three neuropils
    - *neuprint_PNallinrvt_df.pkl*: a pickled instance of Pandas DataFrame containing information about all queried neurons
    - *conn_(PNKC/PNLH)_df.pkl*: a pickled instance of Pandas DataFrame containing the connectivity information
    - *neuron_(PNKC/PNLH)_df.pkl*: a pickled instance of Pandas DataFrame containing the information about third-order neurons
    - *conn_(PNKC/PNLH)_df_8.pkl*: a pickled instance of Pandas DataFrame containing the connectivity information when using higher synaptic weight threshold
    - *neuron_(PNKC/PNLH)_df_8.pkl*: a pickled instance of Pandas DataFrame containing the information about third-order neurons when using higher synaptic weight threshold