Fall 2019 - Journal Club

Schedule

Week of January 13 - Matthew Andres Moreno

Long Term Evolution Experiments!

Langdon, W. B., & Banzhaf, W. (2019). Continuous Long-Term Evolution of Genetic Programming. The 2019 Conference on Artificial Life, 388–395. https://doi.org/10.1162/isal_a_00191

Leon, Dacia et al. “Innovation in an E. coli evolution experiment is contingent on maintaining adaptive potential until competition subsides.” PLoS genetics vol. 14,4 e1007348. 12 Apr. 2018, doi:10.1371/journal.pgen.1007348

Week of January 31 - Alex Lalejini

Paper: S. Kriegman, D. Blackiston, M. Levin, and J. Bongard, “A scalable pipeline for designing reconfigurable organisms,” Proc Natl Acad Sci USA, p. 201910837, Jan. 2020, doi: 10.1073/pnas.1910837117.

Overview

• Authors demonstrate a scalable pipeline for creating functional lifeforms:
  1. use AI methods to automatically design candidate lifeforms in silico to perform a desired function
  2. feasible designs are then used as blueprints for a cell-based construction toolkit
• Authors deployed their pipeline to create systems that exhibited four different behaviors: locomotion, object manipulation, object transport, and collective behavior.

Some Background

Existing methods for designing and building ‘designer’ living systems:

• single-cell that have been genetically engineered
• synthetic organoids can be made by exposing cells to specific culture conditions (there is limited control over structure and function)
• bioengineering with 3D scaffolds

The Pipeline

1. Generator: evolutionary algorithm that discovers different ways of combining the biological building blocks together to realize a desired behavior
2. Build filter:
   • Only designs that are noise-resistant are used as candidates for transfer to cell-based construction
   • Is this design manufacturable?
   • Is this design scalable?
3. Build designs out of frog stem cells
   • Cells are harvested from frog embryos, dissociated, and pooled to achieve a desired number of cells
   • Cells are incubated
   • Cell aggregates are manually shaped by subtraction => microsurgery!
   • Contractile tissue can be embedded/layered into cell aggregate
   • Contractile tissue = cardiac progenitor cells
4. Result: 3D approximation of evolved design, can exist in aqueous environment for period of days to weeks
5. Feed real life results back into initial evolutionary algorithm, repeat.

Questions

• This was an interesting future direction worth talking about: “If equipped with electrically active cells and selected for cognitive or computational functions (30), such designed systems may similarly broaden our under- standing of how intelligence can be instantiated in living as well as nonliving systems.”
• Has anyone had any experience with the cube-y robot system they evolved designs in?
• Radical applications for ‘xenobots’ that were not mentioned in the paper?
• Has anyone looked at their: https://github.com/skriegman/reconfigurable_organisms ?

Week of February 3 - Recent data scandals/RCR

• Social spiders
• Image forensics person

Week of February 10 - Synthetic Ecology

Alex

posted ALife deadline Thursday, 13 February 2020

Week of February 17 - TBD

Week of March 2 - bye

:sunglasses: SPRING BREAK :sunglasses:

Week of March 9 - TBD

Week of March 16 - TBD

Week of March 23 - TBD

Week of March 30 - TBD

Week of April 6 - TBD

Week of April 13 - TBD

Week of April 20 - TBD

Week of April 27 - TBD

:bowing_woman: FINALS WEEK :bowing_man: