Top Banner
Complex Genetic Complex Genetic Evolution of Self- Evolution of Self- Replicating Loops Replicating Loops Chris Salzberg Chris Salzberg 1,2 1,2 Antony Antony Antony Antony 3 Hiroki Sayama Hiroki Sayama 1 1 1 University of Electro-Communications, Japan University of Electro-Communications, Japan 2 2 University of Tokyo, Japan University of Tokyo, Japan 3 3 University of Amsterdam, the Netherlands University of Amsterdam, the Netherlands [email protected] [email protected]
22

Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

Apr 01, 2015

Download

Documents

Emely Clare
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

Complex Genetic Complex Genetic Evolution of Self-Evolution of Self-Replicating LoopsReplicating Loops

Chris SalzbergChris Salzberg1,21,2 Antony Antony Antony Antony33 Hiroki Sayama Hiroki Sayama11

1 1 University of Electro-Communications, JapanUniversity of Electro-Communications, Japan2 2 University of Tokyo, JapanUniversity of Tokyo, Japan

3 3 University of Amsterdam, the NetherlandsUniversity of Amsterdam, the [email protected]@cx.hc.uec.ac.jp

Page 2: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

2

SummarySummary We re-examined the evolutionary We re-examined the evolutionary

dynamics of self-replicating loops on dynamics of self-replicating loops on CA, by using new tools for complete CA, by using new tools for complete genetic identification and genealogy genetic identification and genealogy tracingtracing

We found in the loop populations:We found in the loop populations:1.1. Diversities in macro-scale morphologies Diversities in macro-scale morphologies

and mutational biasesand mutational biases

2.2. Genetic adaptationGenetic adaptation

3.3. Genetic diversification and continuing Genetic diversification and continuing explorationexploration

Page 3: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

3

Background: CA-based AlifeBackground: CA-based Alife

Universal constructor (Von Neumann Universal constructor (Von Neumann 1966; Codd 1968; Takahashi et al. 1990; 1966; Codd 1968; Takahashi et al. 1990; Pesavento 1995)Pesavento 1995)

Self-replicating loops (Langton 1984; Byl Self-replicating loops (Langton 1984; Byl 1989; Reggia et al. 1993)1989; Reggia et al. 1993)

Self-inspecting loops/worms (Ibanez et Self-inspecting loops/worms (Ibanez et al. 1995; Morita et al. 1995, 1996)al. 1995; Morita et al. 1995, 1996)

Self-replicating loops with additional Self-replicating loops with additional capabilities of construction/computation capabilities of construction/computation (Tempesti 1995; Perrier et al. 1996; (Tempesti 1995; Perrier et al. 1996; Chou et al. 1998)Chou et al. 1998)

Spontaneous emergence and evolution Spontaneous emergence and evolution of self-replicators (Lohn et al. 1995; of self-replicators (Lohn et al. 1995; Chou et al. 1997; Sayama 1998, 2000, Chou et al. 1997; Sayama 1998, 2000, 2003; Salzberg et al. 2003, 2004; Suzuki 2003; Salzberg et al. 2003, 2004; Suzuki et al. 2003, 2004)et al. 2003, 2004)

Page 4: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

4

Supposedly Limited Supposedly Limited Evolutionary Dynamics in Evolutionary Dynamics in

CACA McMullin (2000):McMullin (2000):

““[SR loop] does not embody anything like a [SR loop] does not embody anything like a general constructive automaton and general constructive automaton and therefore therefore has little or no evolutionary has little or no evolutionary potentialpotential.”.”

Suzuki et al. (2003):Suzuki et al. (2003):““Though there are many other variations of Though there are many other variations of

CA models for self-replication, CA models for self-replication, their their evolvability does not differ very muchevolvability does not differ very much.”.”

Page 5: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

5

QuestionQuestion

Did we truly understand what was Did we truly understand what was going on in this seemingly simple going on in this seemingly simple dynamics of our CA-based evolutionary dynamics of our CA-based evolutionary systems?systems?

We didn’t know we didn’t, until we We didn’t know we didn’t, until we have developed the formal framework have developed the formal framework and the sophisticated tools for detailed and the sophisticated tools for detailed analysis and visualization for those analysis and visualization for those systems.systems.

Page 6: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

6

Subject: EvoloopSubject: Evoloop An evolvable SR An evolvable SR

loop by Sayama loop by Sayama (1999) constructed (1999) constructed on on nine-state five-nine-state five-neighbor fully neighbor fully deterministic CAdeterministic CA

Robust state-Robust state-transition rules transition rules give rise to give rise to evolutionary evolutionary behaviorbehavior

Mutation/selection Mutation/selection mechanisms are mechanisms are totally totally emergentemergent

Page 7: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

7

New Tools for Detailed New Tools for Detailed AnalysisAnalysis

At every birth, the newborn loop’s At every birth, the newborn loop’s genotype & genotype & phenotypephenotype and its and its genealogical informationgenealogical information is is detected and recorded in an event-driven fashiondetected and recorded in an event-driven fashion

Each genotype-phenotype pair is indexed in the Each genotype-phenotype pair is indexed in the Species DatabaseSpecies Database

genotypegenotype

G G G G C G C G T T G CC CC G

phenotypephenotype88

88

Page 8: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

8

Observation (1):Observation (1):Diversities in Macro-Scale Diversities in Macro-Scale

Morphologies and Morphologies and Mutational BiasesMutational Biases

Page 9: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

9

Huge Genetic State-Huge Genetic State-SpaceSpace

Permutation of genes (Permutation of genes (GG, , TT) and core ) and core states (states (CC) under constraints ) under constraints estimates estimates the number of viable the number of viable genotypesgenotypes to be to be

Size Size nn

# of species# of species Size Size nn

# of species# of species Size Size nn

# of species# of species

44 1515 99 11,44011,440 1414 9,657,7009,657,700

55 5656 1010 43,75843,758 1515 37,442,16037,442,160

66 210210 1111 167,960167,960 1616 145,422,675145,422,675

77 792792 1212 646,646646,646 1717 565,722,720565,722,720

88 3,0033,003 1313 2,496,1442,496,144 1818 2,203,961,42,203,961,43030

22nn-2-2nn-2-2

Page 10: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

10

Diversity in Growth Patterns Diversity in Growth Patterns (size-4)(size-4)

Page 11: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

11

Diversity in Growth Patterns Diversity in Growth Patterns (size-6)(size-6)

Page 12: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

12

Diversity in Mutational Biases Diversity in Mutational Biases (size-6)(size-6)

(new result not included in paper)

Page 13: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

13

Observation (2):Observation (2):Genetic AdaptationGenetic Adaptation

Page 14: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

14

Two Measures of (Possible) Two Measures of (Possible) FitnessFitness

Survival rate Survival rate (sustainability in (sustainability in competition):competition):— Characterized by an average of relative Characterized by an average of relative

population ratios of a species after a given population ratios of a species after a given period of time in competition with another period of time in competition with another speciesspecies

Colony density index Colony density index (growth speed):(growth speed):— Characterized by a quadratic coefficient of Characterized by a quadratic coefficient of

a parabola fitted to the population growth a parabola fitted to the population growth curve of each species in an infinite domaincurve of each species in an infinite domain

Page 15: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

15

Variety and Correlation Variety and Correlation (size-4)(size-4)

Page 16: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

16

Evolution Evolution in vivoin vivo(starting from size-8)(starting from size-8)

Page 17: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

17

Evolution Optimizes Evolution Optimizes “Fitness”“Fitness”

Evolutionary transitionactually observed inthe previous slide

Page 18: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

18

Observation (3):Observation (3):Genetic Diversification and Genetic Diversification and

Continuing ExplorationContinuing Exploration

Page 19: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

19

Non-Mutable SubsequencesNon-Mutable Subsequences

Certain subsequences are found Certain subsequences are found non-mutablenon-mutable::G{G{C*C*}}TT{{C*C*}}TTGG

A long non-mutable sub-sequence injected to A long non-mutable sub-sequence injected to ancestor causes a ancestor causes a relatively largerelatively large lower lower bound of viable sizesbound of viable sizes upon its descendants, a upon its descendants, a reduced size-based selection pressurereduced size-based selection pressure, and a , and a highly biased highly biased mutational tendency to larger mutational tendency to larger speciesspecies

Such “GMO” loops show long-lasting Such “GMO” loops show long-lasting evolutionary exploration processesevolutionary exploration processes

GGGGGGGGCCGGC C GGCCCCTTCCCCTTG GG G

Page 20: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

20

control

with long non-mutable subsequences

with subsequences + hostile environment

(new result not included in paper)

Page 21: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

21

ConclusionsConclusions

Huge diversityHuge diversity, , non-trivial genetic non-trivial genetic adaptationadaptation and and diversificationdiversification unveiled unveiled in the evoloop systemin the evoloop system

Hierarchical emergence Hierarchical emergence demonstrated, where demonstrated, where macro-scale macro-scale evolutionary changes of populations evolutionary changes of populations arises from micro-scale interactions arises from micro-scale interactions between elements much smaller than between elements much smaller than individual replicatorsindividual replicators, traversing , traversing multiple scalesmultiple scales

Page 22: Complex Genetic Evolution of Self-Replicating Loops Chris Salzberg 1,2 Antony Antony 3 Hiroki Sayama 1 1 University of Electro-Communications, Japan 2.

22

References & AcksReferences & Acks Salzberg, C. (2003) Salzberg, C. (2003) Emergent Evolutionary Dynamics of Emergent Evolutionary Dynamics of

Self-Reproducing Cellular AutomataSelf-Reproducing Cellular Automata. M.Sc. Thesis. . M.Sc. Thesis. Universiteit van Amsterdam, the Netherlands.Universiteit van Amsterdam, the Netherlands.

Salzberg, C., Antony, A. & Sayama, H. Visualizing Salzberg, C., Antony, A. & Sayama, H. Visualizing evolutionary dynamics of self-replicators: A graph-based evolutionary dynamics of self-replicators: A graph-based approach. approach. Artificial LifeArtificial Life, in press., in press.

Sayama, H. Sayama, H. The SDSR loop / Evoloop Homepage.The SDSR loop / Evoloop Homepage. http://complex.hc.uec.ac.jp/sayama/sdsr/http://complex.hc.uec.ac.jp/sayama/sdsr/

Antony, A. & Salzberg, C. Antony, A. & Salzberg, C. The Artis Project Homepage.The Artis Project Homepage. http://artis.phenome.org/http://artis.phenome.org/

This work is supported in part by the Hayao Nakayama Foundation for Science, This work is supported in part by the Hayao Nakayama Foundation for Science, Technology & Culture, and the International Information Science Foundation, Technology & Culture, and the International Information Science Foundation, Japan.Japan.