A Connectome and Analysis of the Adult Drosophila Central Brain

Louis K. Scheffer, C. Shan Xu, Michał Januszewski, Zhiyuan Lu, Shin-ya Takemura, Kenneth J. Hayworth, Gary B. Huang, Kazunori Shinomiya, Jeremy Maitin-Shepard, Stuart Berg, Jody Clements, Philip M. Hubbard, William T. Katz, Lowell Umayam, Ting Zhao, David Ackerman, Tim Blakely, John Bogovic, Tom Dolafi, Dagmar Kainmueller, Takashi Kawase, Khaled Khairy, Laramie Leavitt, Peter H. Li, Larry Lindsey, Nicole Neubarth, Donald J. Olbris, Hideo Otsuna, Eric T. Trautman, Masayoshi Ito, Jens Goldammer, Tanya Wolff, Robert Svirskas, Philipp Schlegel, Erika Neace, Yinan Wan, Chelsea X. Alvarado, Dennis Bailey, Samantha Ballinger, J Borycz, Brandon S Canino, Natasha Cheatham, Michael Cook, Marisa Dreher, Octave Duclos, Bryon Eubanks, Kelli Fairbanks, Samantha Finley-May, Nora Forknall, Audrey Francis, Gary Patrick Hopkins, Emily Joyce, Sungjin Kim, Nicole Kirk, Julie Kovalyak, Shirley A Lauchie, Alanna Lohff, Charli Maldonado, Emily A Manley, Sari McLin, Caroline Mooney, Miatta Ndama, Omotara Ogundeyi, Nneoma Okeoma, Christopher Ordish, Nicholas Padilla, Christopher Patrick, Tyler Paterson, Elliott Phillips, Emily M Phillips, Neha Rampally, Caitlin Ribeiro, Madelaine Robertson, Jon Thomson Rymer, Sean M Ryan, Megan Sammons, Anne K Scott, Ashley L Scott, Aya Shinomiya, Claire Smith, Kelsey Smith, Natalie Smith, Margaret A. Sobeski, Alia Suleiman, Jackie Swift, Satoko Takemura, Iris Talebi, Dorota Tarnogorska, Emily Tenshaw, Temour Tokhi, John J Walsh, Tansy Yang, Jane Anne Horne, Feng Li, Ruchi Parekh, Patricia K. Rivlin, Vivek Jayaraman, Kei Ito, Stephan Saalfeld, Reed GeorgebioRxiv 2020 · DOI

TL;DR

This paper presents the first comprehensive connectome of the adult Drosophila central brain, detailing the intricate neural circuitry and providing a public resource for future research. By developing novel methods for electron microscopy, image processing, and data analysis, the authors have mapped ~25,000 neurons and ~20 million synapses, offering unprecedented insights into brain organization and function.

What Problem Does This Paper Address?

Understanding the neural circuits that underpin animal behavior is a fundamental challenge in neuroscience. While Drosophila has been a powerful model organism, a complete, high-resolution map of its central brain's neural connections (connectome) has been lacking. Previous efforts were either limited in scope, resolution, or accessibility. This paper addresses this gap by providing a dense, detailed reconstruction of a significant portion of the adult Drosophila central brain, enabling a more holistic understanding of neural computation and behavior.

Methods

The authors employed a multi-stage approach involving significant methodological advancements:

  • Sample Preparation: Novel methods for fixing, staining, and embedding fly brains were developed to preserve ultrastructural detail and enhance contrast for electron microscopy.
  • Imaging: Advanced Focused Ion Beam Scanning Electron Microscopy (FIB-SEM) was used to acquire high-resolution, isotropic image data. To manage the large volume, the brain was sectioned into 20 µm thick slabs, which were then imaged.
  • Image Processing: A sophisticated pipeline involving image alignment (using custom algorithms and CycleGANs for normalization), automated segmentation (using flood-filling networks), and synapse prediction (employing machine learning) was developed.
  • Proofreading: A crucial, labor-intensive process involving human experts to correct segmentation and synapse prediction errors was implemented, utilizing novel visualization tools.
  • Data Analysis: The reconstructed data was used to define brain regions, classify cell types based on morphology and connectivity, and analyze circuit properties like connection strength distributions and neural motifs.
  • Data Accessibility: A web-based browser (NeuPrint) and APIs were developed to make the massive dataset publicly accessible and queryable.

Datasets

  • Training Data: Dense synapse annotations within small cubes (~122 cubes) were collected for training and validation of machine learning models. This included ~7.6k T-bars and ~11.7k PSDs.
  • New Benchmarks: The paper implicitly introduces a new benchmark by providing the most comprehensive Drosophila central brain connectome to date. The publicly available dataset and analysis tools serve as a foundation for future comparative studies and validation efforts.

Key Insights and Ideas

  • Packing Density as an Evolutionary Driver: The study provides evidence suggesting that maximizing packing density is a significant evolutionary criterion in brain evolution, observed in both biological and human-designed systems.
  • Short Paths and Tight Coupling: Drosophila brain neurons are highly interconnected, with short path lengths between neuron pairs, contrasting with longer path lengths in human-designed computational systems.
  • Power Law Distribution of Connection Strengths: Unlike the log-normal distribution found in mammalian cortex, Drosophila central brain connection strengths follow a power law with an exponential cutoff, suggesting different principles of neural circuit organization.
  • Ubiquity of Reciprocal Connections: Reciprocal connections are a common motif across all brain regions and cell types in the fly brain, indicating a fundamental feature of neural circuit design.
  • Novel Cell Types and Circuits: The comprehensive mapping revealed many novel cell types and circuits, highlighting the complexity of the Drosophila brain beyond previously studied regions.

Key Results

  • Comprehensive Connectome: The paper presents a dense reconstruction of approximately 25,000 neurons and ~20 million chemical synapses in the adult Drosophila central brain.
  • Cell Type Classification: Over 18,000 neuronal cell bodies were identified, leading to the classification of ~4,890 distinct cell types.
  • Synapse Prediction Accuracy: Achieved high precision and recall rates for synapse prediction, with end-to-end synapse prediction accuracy reaching 88% median precision.
  • Data Accessibility: The entire dataset is publicly available through the NeuPrint web interface, enabling broad scientific access.

Limitations and Future Work

  • Hemibrain Coverage: The dataset covers only a portion of the Drosophila brain (primarily the right hemisphere), with some regions missing or incompletely reconstructed.
  • Sex Differences: The current connectome is from a single female fly, and differences between sexes are known to exist and warrant future investigation.
  • VNC and Optic Lobes: The ventral nerve cord (VNC) and optic lobes are not fully included, limiting the analysis of motor control and visual processing circuits.
  • Future Work: The authors propose several future directions, including reconstructing the male brain, completing the VNC and optic lobes, imaging the entire fly brain, and further exploring the functional implications of the connectome, such as understanding the circuits underlying behaviors that are not yet identified. They also emphasize the need for continued refinement of reconstruction and analysis methods.

Summary updated

Abstract

Abstract The neural circuits responsible for animal behavior remain largely unknown. We summarize new methods and present the circuitry of a large fraction of the brain of the fruit fly Drosophila melanogaster. Improved methods include new procedures to prepare, image, align, segment, find synapses in, and proofread such large data sets. We define cell types, refine computational compartments, and provide an exhaustive atlas of cell examples and types, many of them novel. We provide detailed circuits consisting of neurons and their chemical synapses for most of the central brain. We make the data public and simplify access, reducing the effort needed to answer circuit questions, and provide procedures linking the neurons defined by our analysis with genetic reagents. Biologically, we examine distributions of connection strengths, neural motifs on different scales, electrical consequences of compartmentalization, and evidence that maximizing packing density is an important criterion in the evolution of the fly’s brain.
BibTeX
@misc{scheffer2020connectome,
  author = {Louis K. Scheffer and C. Shan Xu and Michał Januszewski and Zhiyuan Lu and Shin-ya Takemura and Kenneth J. Hayworth and Gary B. Huang and Kazunori Shinomiya and Jeremy Maitin-Shepard and Stuart Berg and Jody Clements and Philip M. Hubbard and William T. Katz and Lowell Umayam and Ting Zhao and David Ackerman and Tim Blakely and John Bogovic and Tom Dolafi and Dagmar Kainmueller and Takashi Kawase and Khaled Khairy and Laramie Leavitt and Peter H. Li and Larry Lindsey and Nicole Neubarth and Donald J. Olbris and Hideo Otsuna and Eric T. Trautman and Masayoshi Ito and Jens Goldammer and Tanya Wolff and Robert Svirskas and Philipp Schlegel and Erika Neace and Yinan Wan and Chelsea X. Alvarado and Dennis Bailey and Samantha Ballinger and J Borycz and Brandon S Canino and Natasha Cheatham and Michael Cook and Marisa Dreher and Octave Duclos and Bryon Eubanks and Kelli Fairbanks and Samantha Finley-May and Nora Forknall and Audrey Francis and Gary Patrick Hopkins and Emily Joyce and Sungjin Kim and Nicole Kirk and Julie Kovalyak and Shirley A Lauchie and Alanna Lohff and Charli Maldonado and Emily A Manley and Sari McLin and Caroline Mooney and Miatta Ndama and Omotara Ogundeyi and Nneoma Okeoma and Christopher Ordish and Nicholas Padilla and Christopher Patrick and Tyler Paterson and Elliott Phillips and Emily M Phillips and Neha Rampally and Caitlin Ribeiro and Madelaine Robertson and Jon Thomson Rymer and Sean M Ryan and Megan Sammons and Anne K Scott and Ashley L Scott and Aya Shinomiya and Claire Smith and Kelsey Smith and Natalie Smith and Margaret A. Sobeski and Alia Suleiman and Jackie Swift and Satoko Takemura and Iris Talebi and Dorota Tarnogorska and Emily Tenshaw and Temour Tokhi and John J Walsh and Tansy Yang and Jane Anne Horne and Feng Li and Ruchi Parekh and Patricia K. Rivlin and Vivek Jayaraman and Kei Ito and Stephan Saalfeld and Reed George},
  title = {{A Connectome and Analysis of the Adult <i>Drosophila</i> Central Brain}},
  year = {2020},
  doi = {10.1101/2020.04.07.030213},
  howpublished = {bioRxiv},
}

Loading related content...

Figure