*Genetic Algorithms
and their Application to the Artificial Evolution of Genetic Regulatory Networks*

Overview

*Part 1: Fundamentals*

Biological Evolution

Biological Evolution - Example

Biological Evolution – cont‘d

Evolutionary cycle : Generation

Dictionary 1

Selection and Reproduction

Recombination (crossover)

Duplication

Mutation

The Evolutionary Mechanisms

*Part 1: Fundamentals (2)*

Evolutionary Computation

Evolutionary Computation

Optimization

Optimization - Problems

Optimization - Problems

Optimization - Problems

Evolutionary cycle - revisited

Dictionary 2

EC - General properties

Main trends

Genetic Algorithms

GAs - Simple Example

Simple example – f(x) = x²

F(x) = x² - Start Population

F(x) = x² - Selection

F(x) = x² - Selection

F(x) = x² - Selection

F(x) = x² - Recombination

F(x) = x² - Recombination

F(x) = x² - Mutation

F(x) = x²

F(x) = x² - End

F(x) = x² - Animation

GAs - General

Genetic algorithms

History

Coding and Mapping

Genetic coding

Genetic coding and mapping

Mapping – Example

Mapping – Example cont‘d

Mapping – Example cont‘d

Mapping – Example cont‘d

Mapping – Example cont‘d

Mapping – Example cont‘d

Selection

Selection

Selection - Roulette-Wheel

Selection - Elitism

Selection - Tournament

Genetic variability operators

Mutation

Recombination/Crossover

Recombination

Fitness

Fitness function

Fitness space or landscape

Fitness space or landscape

Fitness landscapes contd.

Fitness space – Good design

Fitness space - Bad design

Fitness space – Mediocre design

Dynamic fitness landscape

Design issues

Integrating problem knowledge

Design decisions

Hints for the parameter search

Keep in mind

Wrapping up Part 1

GA- Summary

GA- Summary cont'd

More recent inputs from Biology

*Part 2: Modelling GRNs*

Evolutionary cycle - again

Evolutionary cycle - again

Selection

Decoding, Evaluating, Comparing

Decoding

Decoding

Evaluating

Comparing

Decoding, Evaluating, Comparing

Decoding, Evaluating, Comparing

Decoding, Evaluating, Comparing

(Artificial) Genetic Regulatory Networks (aGRN)

GRNs and aGRNs

GRN-theory – Central Dogma

GRN-theory – Central Dogma

GRN Control

Control of gene expression

Control of gene expression

Control of gene expression

Control of gene expression

GRN Dynamics

Dynamics: Petri-net notation

Dynamics: Petri-net notation

Symbols: Nodes

Symbols: Arrows

Dynamics: what happens?

Dynamics: what happens?

Dynamics: what happens?

Dynamics: what happens?

Dynamics: what happens?

Dynamics: what happens?

Dynamics: what happens?

Dynamics: what happens?

Worth remembering (1)

Dynamics

Dynamics

Worth remembering (2)

Combining control and dynamics

Creating a dynamic GRN model - Questions (1)

Creating a dynamic GRN model – Questions (2)

Basic GRN model

Possible dynamic representation

Rate equations

Rules for activation and repression

Rules for combinatorial modifier effects

Note on GRN representation

*Part 3: Evolving biological clocks*

Introduction

Schematic drawing

Schematic drawing - Phenotype

What does a gene-node consist of?

Encoding of gene-nodes and genome

Encoding of gene-nodes and genome – cont’d

Activation

Activation – cont’d

Gene-node activation example

Environmental in-/output

Selection and Variation

Recombination

Population development example

Individual dynamics analysis

Evolvability - Heterochronicity

Differentiation – schematic

Differentiation: two pathways

Typical example runs for the two settings

*Part 4: Netbuilder'*

NetBuilder′ (Project Apostrophe)

NetBuilder′ NetBuilder

NetBuilder′ - Petri net

NetBuilder′ - GUI

NetBuilder′ - GUI

NetBuilder′ - GUI

NetBuilder′ - GUI

NetBuilder′ - GUI

Evolution in NetBuilder'

Genotype - Phenotype

Genotype – Phenotype (2)

Furthermore

Selection

Recombination

Mutations

Fitness

Parameters

Parameters

Parameters

Summary – NetBuilder'

ICSB 2007 - Posters

Acknowledgement

Resources –
Evolutionary Algorithms

References –
Evolving biological clocks with aGRNs