/Connectome-Cloning

Connectome cloning project.

Primary LanguageMATLAB

Connectome Cloning

This document countains details regarding the connectome cloning project.

Tasks

Here are some finalizing tasks.

  • Flickering synapses (15% of all available synapses are present)

Discussion

For flickering synapses, since both swapping and flipping is direction agnostic, the flicker map must be symmetric.

Workflow

The function connclone is the latest version of the algorithm. That function can be checked to see the latest format to enter data. All scripts in this directory are wrappers to record and save data. Graph generating scripts are put in visuals/generators.

Functions

Besides the main function, there are multiple modules.

  • Flicker: Simulation with flickering synapses, meaning the connection is not TR, but some random connectivity, with changes to this connectivity happening at bulk in preset intervals.
  • Parallel: In this simulation, all barcodes try to move simultaneously.

Variables

Free letters remaining: G

These are variables used to keep track of things. Originally on connclone, these might differ between versions. Will add version spoecific ones here as well.

Inputs

  • N 100: Input connectome size
  • D .1 : Connectome density
  • E 10 : Cost parameter
  • H 0,0: Temperature scaling.
  • fd .3: Active synapse percentage (flickering only).

Outputs

  • O: 0 if reconstruction was unsuccessful, convergence time if successful.
  • B: number of barcode pairs in the connectome

Runtime

  • W: Used for batch generating random numbers.
  • G: Sets the maximum number of times the simulation is allowed to run (in average runtime)
  • C: The original connectome
  • V: Reconstructed connectome
  • T: Max simulation runtime
  • R: Move probabilities
  • L: Energy (above GS)

Flicker: for changing activation of synapses

  • P: Number of times the synapse activation should change.
  • flicker_step: Step size for flickering synapses.

State

  • K: List of barcodes on bp ends (pre,pos)
  • S: List that contains bp locations. (pre,pos)

Tracking

For hamiltonian;

  • M: Table keeping track of barcodes per cell (barc,cell)
  • F: Total barcodes per cell (cell)

For cell selection

  • A: List of bp ids in each cell (cell, ...)
  • J: Total # of barcode pairs per cell (cell)
  • Q: Location of bp in A (cell,bp-id)

For swap candidate selection

  • Y: List of cells with >1 bp in them (...)
  • Z: Total # of cells with >1 bp in them
  • X: Location of cells in A (cell)

Flicker: for jump move selection

  • I: List of synapses to cells. (cell, ...)
  • U: Total number of synapses per cell