Temporal and Notch identity determine layer targeting and synapse location of medulla neurons I. Holguera, Y-C. Chen, Y-C-D. Chen, F. Simon, A.G. Gaffney, J.D. Rodas, S. Córdoba, C. Desplan You can find the manuscript associated with the analysis in preprint.
This repository analyzes how connectivity in the fly visual system is associated with their developmental origin using FlyWire connectome.
The analysis pipeline consists of four main modules:
- Filters out unreliable connections with fewer than 5 synapses
- Labels each connection with the cell types involved
- Splits data by neuropil for efficient processing
- Rotates 3D coordinates to a Fischbach-esque orientation
- Compare how the depths of synapses vary with developmental origin and function
- Tests if neurons born at similar times tend to have similar functions
- Make scatter plots showing where synapses are located in a Fischbach-esque perspective
- Color connections by cell type, birth time, or function
- Create trees showing which cell types are most similar in their connectivity patterns
- Generates donut charts showing the partners of neuronal types of interest
- Highlights synapses from neuron expressing specific genes
- Tests terminal selector genes for superficial vs deep depth bias in ME_R, LO_R, and LOP_R neuropils
- Runs neuron-level bootstrap analysis for both presynaptic and postsynaptic compartments
- Outputs FDR-corrected statistics with combined CSV and Excel summaries
This pipeline has been tested on NYU HPC running Red Hat Enterprise Linux 10.1 (Coughlan)). To reproduce the analysis, you will need:
Installation of the below dependencies should take around an hour on a personal machine, but given the size of the data, preprocessing and statistics would take a long time to run. Therefore it is recommended to be used on an HPC environment where tasks can be divided and run in parallel.
System dependencies are distributed via the official websites provided below, or via your preferred package/dependency manager.
- Nextflow (workflow manager; v25.10.2; https://nextflow.io/)
- R (version 4.0 or higher; https://cran.r-project.org/)
- C++ compiler (for statistical modules; C++11 or above; https://gcc.gnu.org/)
- Access to FlyWire dataset files (https://codex.flywire.ai)
- data.table
- ggplot2
- Rcpp
- openxlsx
- dplyr
- patchwork
- cowplot
- ggdendro
# Run specific modules
nextflow run src/preprocessing/main.nf
nextflow run src/stats/main.nf
nextflow run src/visualize/main.nf
nextflow run src/selector_test/main.nfThe connectome data used in this study were generated by the FlyWire Consortium and are distributed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).
The demo subset included in this repository is redistributed under the same license with attribution as required. When using or citing this data, please follow the FlyWire citation guidelines and cite the applicable primary papers:
- Dorkenwald, S. et al. Neuronal wiring diagram of an adult brain. Nature 634, 124–138 (2024).
- Schlegel, P. et al. Whole-brain annotation and multi-connectome cell type quantification. Nature 634, 139–152 (2024).
- Matsliah, A., Yu, S.-c. et al. Neuronal parts list and wiring diagram for a visual system. Nature 634, 166–180 (2024).
- Brain connectivity data: FlyWire connectome dataset (Dorkenwald et al., Nature 2024)
- Cell type classifications: Visual neuron annotations from FlyWire consortium
- Similarity measurements: Connectivity dissimilarity matrix from Matsliah et al., Nature 2024
- Binarized gene expression: Transcription factor and cell adhesion molecule patterns (Özel et al., Nature 2020)
- Demo data: A subset data from FlyWire for demo runs to test the codebase.
For detailed information about each module, see the individual README files in the src/ subdirectories.
For top-level manual/helper scripts in src/, see src/README.md.