In the right optic lobe of the FlyWire Connectome there are 23,452 "columnar" neurons of 31 distinct types that were divided into 796 "columns" by combining manual work and computational optimization. Each column consists of a group of neurons that are responsible for processing visual sensory information from a specific region of the visual field. Grouping cells into columns provides an important abstraction for reasoning about visual circuitry. On this page you can view the current "best" column assignment, and the goal of this challenge is to optimize it, by increasing the fraction of connections (synapses) inside the columns (vs cross columns).

Prerequisites

This is a data analysis / optimization challenge, and it is open to everyone. Participants do not have to know anything about the Drosophila brain or connectomics. Data processing and some algorithmic skills should suffice to compete.

Constraints

• Each cell type can have at most one cell in every column. In other words, columns should contain at most one cell per type.
• There are 796 columns numbered 1-796. These are the only available columns (no new columns can be created).
• Some cell types contain less cells than the number of columns, and some contain more. You are free to choose which cells to assign to which columns, and leave any number of cells "unassigned".

Data

• Connectivity between cells with synapse counts (download here)
• Cell types and current best assignment to columns (download here)
• General metadata (download here)

Objectives

• The quality of the assignment to columns is measured in terms of the number of synapses between cells in same column. In the current "benchmark" assignment 1,381,915 synapses are such.
• Secondary objective is to maximize the number of cells assigned to columns. Currently 22,578 cells are assigned and the rest (874) are not assigned.
• Improvement in any of these metrics (without degrading the other metric) will qualify for a spot in the leaderboard.

How to submit solutions

• Email your solution to arie@princeton.edu, including:
  • CSV file with updated assignment to columns (same format as this)
  • Your score - number of synapses within columns
  • Short explanation of how the improvement was achieved (method / main idea)
  • Mention if it is ok to have your name displayed on the leaderboard and how
• Submissions will be verified and leaderboard updated periodically, until end of April 2024.
• There is no limit on number of submissions per participant. Only the best one will be shown.
• Winners of the challenge will receive a special plaque from the FlyWire team at Princeton University, and will be invited to give a short presentation (optional).