Class Notes (1,000,000)
CA (610,000)
Western (60,000)
BIOL (7,000)
BIOL 1002B (1,000)
Lecture

Biology 1002B Lecture Notes - Pink Noise, White Noise, Complex System


Department
Biology
Course Code
BIOL 1002B
Professor
Denis Maxwell

Page:
of 2
Lecture 25: Life as a Complex System
1. Explain the different colours of reaction in the microarray of Fig 15.20.
Normal/control cell use nucleotides with green fluorescent
Cancer/experimental cell use nucleotides with red fluorescent
Coloured spots on microarray indicate where labelled cDNAs bound to gene probes attached to chip
o Which genes were active in normal and/or cancer cells
o Pure green active in normal but not cancer
o Pure red active in cancer but not normal
o Yellow active in both cells
o Other colours tells us relative levels of gene expression
2. Summarize characteristics of complex systems (from TedX talk at left).
Game Theory success of one person depends on the choices of another person
Complex systems outcome is more elaborate than the sum of its parts
3. Direct causal
Interaction of subgroup of agents can cause
observable pattern
Interaction of some agents might have
central controlling effects on the pattern
Direct correspondence between interaction
and the pattern
o Interaction = pattern
Some interaction are “intentional” to
achieve a global goal (pattern)
Pattern arises from additive summing of
particular sequential events
E.g. sequential story
Not causal or emergent white noise (equal
amount of all frequencies in a range shows
flat line; no relationship)
o Pink noise is more ordered and not
random
Emergent Explanatory Schema
Interaction among entire collection of
agents “causes” pattern
Interaction among agents give rise to
decentralized or distributed control
Interactions among agents and overall
pattern are not necessarily aligned
Agents interact to achieve local goals only
o Global pattern emerges from
undirected local interactions
Patterns arise from net effect of all
interactions
Characteristics of the whole are not
predictable by knowledge of the parts
Dynamic attractors emerge on boundary
between high order and high disorder
E.g. complex systems (life, humans), traffic
jams, Zipf’s Law of language (frequency of
any word is inversely proportional to its
rank in frequency table), earthquakes,
healthy heart rate
E.g. pink noise sound where short waves/
high pitch high proportion
4. How molecules behave when diffusing in water
Emergent
Not directional but random
Distribution is only at one point of time and does not stop once distributed
o Gradient is not the driving force since ATP moves because of thermal energy
5. Complicated
Many parts
Defined relationships
Highly controlled
Sequential
High predictability
Complex Systems
Many self-organized parts
Changing relationships
No central control
Emergent patterns (1/f - linear)
o Log frequency vs log magnitude
Low predictability
6. Reasons why large effects do not require large causes in complex systems
Linear straight line graph means they create the same result of large and small cause for same reason
Nothing special or unusual that causes large consequence
Small changes can cause big consequences
Big consequences are not more rare
7. Definition of "attractor state"
Attractor is a state of a system that is relatively stable used to understand complex system
Attractor is a set towards which a dynamical system evolves over time
o Points that get close enough to attractor remain close even if slightly disturbed
E.g. where the marble is the state of the system
o Where the marble always end up (stable location) is the attractor
o The bottom attracts marble/system to the state
E.g. tub swirl is a dynamic attractor
o Water molecules are randomly distributed (highly disordered)
o When you pull the plug, swirl more ordered and less entropy
o Self-organized, emergent, dynamic, transient, emerge on boundary between order and disorder
maintained by constant input of energy (potential energy is gone so need to keep adding water)
Landscape all the states/intersections
o Give rise to attractors
o Hard to go up the hill and go into another bowl
8. Mechanisms of microarrays
Assess expression of thousands of genes in a network/web
o Access gene products that are available at any one time
E.g. EGF network (big web of interaction)
9. Interpretation of microarray data
Gene expression in networks creates a “landscape”
Every intersection in a particular collection of gene products
Move along landscape and end up at decision point
Different phenotypes due to developmental attractor give rise to a stable kind of cell
10. How the location of (or access to) attractor states might be changed on a gene expression landscape
What is cancer, is it a developmental attractor?
What if there is gene duplication, how would landscape be different?
Evolution is sculpting the landscape