Title: Complexity of Life and Intelligent Design By Chance or by Design Betsy Siewert Sept' 16, 2006
1Complexity of Life and Intelligent DesignBy
Chance or by Design?Betsy SiewertSept. 16, 2006
2- Example of Complexity of Life
- Tutorial in Design Inference
- PhD thesis of William Dembski
- Application of Design Inference
- Conclusions
3Eyespot of Chlamydomonas
- Flagella beat 25-50/sec
- Forward 4 um / flagellar beat
- Backward 2 um / flagellar beat
- Rotates 8o/flagellar beat
- Helical pattern of 2/sec
- 100 um / sec
MICROBIOLOGICAL REVIEWS, Dec. 1980, p. 572-630
44 problems to solve in order to swim toward the
light
54 problems to solve in order to swim toward the
light
- Sensitivity to detect light
- Light vs. no light
64 problems to solve in order to swim toward the
light
- Sensitivity to detect light
- Light vs. no light
- Wide range of light intensities
- Less light vs. more light
74 problems to solve in order to swim toward the
light
- Sensitivity to detect light
- Light vs. no light
- Wide range of light intensities
- Less light vs. more light
- Signal to noise ratio
- Diffraction of light from surface of water
- Random rotation of organism
- Convection currents of water
84 problems to solve in order to swim toward the
light
- Sensitivity to detect light
- Light vs. no light
- Wide range of light intensities
- Less light vs. more light
- Signal to noise ratio
- Diffraction of light from surface of water
- Random rotation of organism
- Convection currents of water
- Communication with flagella
9Eyespot of Chlamydomonas
http//web.syr.edu/7Esrsangia/research.html
http//web.syr.edu/srsangia/projects.html
10(No Transcript)
11Decision Tree Using Design Inference
High Probability?
Intermediate Probability?
Small Probability? Specified?
12Decision Tree Using Design Inference
High Probability?
Necessity
yes
Intermediate Probability?
Small Probability? Specified?
13Decision Tree Using Design Inference
High Probability?
Necessity
yes
no
Intermediate Probability?
Small Probability? Specified?
14Decision Tree Using Design Inference
High Probability?
Necessity
yes
no
Intermediate Probability?
Chance
yes
Small Probability? Specified?
15Decision Tree Using Design Inference
High Probability?
Necessity
yes
no
Intermediate Probability?
Chance
yes
no
Small Probability? Specified?
16Decision Tree Using Design Inference
High Probability?
Necessity
yes
no
Intermediate Probability?
Chance
yes
no
Small Probability? Specified?
Design
yes
17Decision Tree Using Design Inference
18Is this Specified?
- HHTTTTHHTHTHHTTTHHTHHHHHHHTHTTTHHTTTHHTHHTTHHHTHH
HTTTHHTTHTTTTHTHHHHTHHHTHHTTHHHHHTHTTHTHTTHTHTHTHH
HTTHHHHHH
19Is this Specified?
- HTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTH
THTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTH
THTHTHTHT
20Example of Using Decision Tree
21Assumptions
- Safe had been securely locked
- Combination lock has 100 numbers
- 00 to 99
- Combination lock requires 5 alternating turns
- Only one specified sequence of 5 numbers will
open the lock
22How many possible combinations are there?
23How many possible combinations are there?
24How many possible combinations are there?
- 100100100100100
- 1005
- 10,000,000,000
- (10 billion)
25Decision Tree Using Design Inference
26(No Transcript)
27Probability
- Definition The probability of an event (E) with
respect to hypothesis of background information
(Ho) is the best available estimate of how likely
E is to occur under the assumption that Ho is
true. - Notation P(EHo)
28If Background information (Ho) changes, then
probability (P) might change
- E Event that John will go to a party on Friday
29If Background information (Ho) changes, then
probability (P) might change
- E Event that John will go to a party on Friday
- Ho1 John absolutely loves parties
30If Background information (Ho) changes, then
probability (P) might change
- E Event that John will go to a party on Friday
- Ho1 John absolutely loves parties
- P(EHo1)?
31If Background information (Ho) changes, then
probability (P) might change
- E Event that John will go to a party on Friday
- Ho1 John absolutely loves parties
- Ho2 John was in a car accident and is in ICU
32If Background information (Ho) changes, then
probability (P) might change
- E Event that John will go to a party on Friday
- Ho1 John absolutely loves parties
- Ho2 John was in a car accident and is in ICU
- P(EHo1, Ho2) lt P(EHo1)
33If Background information (Ho) changes, then
probability (P) might change
- E Event that John will go to a party on Friday
- Ho1 John absolutely loves parties
- P(EHo1)?
34If Background information (Ho) changes, then
probability (P) might change
- E Event that John will go to a party on Friday
- Ho1 John absolutely loves parties
- Ho2 John has blond hair
35If Background information (Ho) changes, then
probability (P) might change
- E Event that John will go to a party on Friday
- Ho1 John absolutely loves parties
- Ho2 John has blond hair
- P(EHo1, Ho2) P(EHo1)
36Complexity
- Definition The complexity of a problem (Q) with
respect to resources (R) is the best available
estimate of how difficult it is to solve Q under
the assumption that R is true. - Notation f(QR)
http//www.math.utah.edu/pa/math/mandelbrot/mande
lbrot.html
37Complexity and Probability
- Complexity and Probability are inversely related
- Probability Complexity
- P(EHo) f (QR)
- 0,1 0, 8)
38Complexity and Probability
- Shannon information is a type of measure of
complexity - I -log2 (p)
39Complexity and Probability
- Shannon information is a type of measure of
complexity - I -log2 (p)
- Bank Safe
- f (Q) -log2 (10-10) 33
40Tractability Bounds for Complexity
- f(QR) lt l Q is tractable
- f(QR) gt m Q is intractable
- Notation (f, l)
41(No Transcript)
42Specificity
- Pattern
- Specified pattern
- Fabricated pattern
- Example
43Specificity
- Pattern
- Specified pattern
- Fabricated pattern
- Example
http//www.fotosearch.com
44Example
- THTTTHHTHHTTTTTHTHTTHHHTTHTHHHTHHHTTTTTTTHTTHTTTHH
THTTTHTHTHHTTHHHHTTTHTTHHTHTHTHHHHTTHHTHHHHTHHHHTT
-
45Example
- THTTTHHTHHTTTTTHTHTTHHHTTHTHHHTHHHTTTTTTTHTTHTTTHH
THTTTHTHTHHTTHHHHTTTHTTHHTHTHTHHHHTTHHTHHHHTHHHHTT
- 01000110110000010100111001011101110000000100100011
01000101011001111000100110101011110011011110111100
46Example
- 01000110110000010100111001011101110000000100100011
01000101011001111000100110101011110011011110111100
47Example
- 01000110110000010100111001011101110000000100100011
01000101011001111000100110101011110011011110111100
- 0 1 00 01 10 11 000 001 010 011 100
101 110 111 0000 0001 0010 0011 0100 0101
0110 0111 1000 1001 1010 1011 1100 1101
1110 1111 00
48Notation for Specificity
- E Set of events All possible results from 100
coin flips
49Notation for Specificity
- E Set of events 100 coin flips
- E One instance of E THTTTHHTHH
50Notation for Specificity
- E Set of events 100 coin flips
- E One instance of E THTTTHHTHH
- D Function that maps a descriptive language D
to the set of possible events E
51Notation for Specificity
- E Set of events 100 coin flips
- E One instance of E THTTTHHTHH
- D Function that maps a descriptive language D
to the set of possible events E - D mapping of one event 010001101
52Notation for Specificity
- E Set of events 100 coin flips
- E One instance of E THTTTHHTHH
- D Function that maps a descriptive language D
to the set of possible events E - D mapping of one event 010001101
- Ho Hypothesis of background information
fair coin
53Notation for Specificity
- E Set of events 100 coin flips
- E One instance of E THTTTHHTHH
- D Function that maps a descriptive language D
to the set of possible events E - D mapping of one event 010001101
- Ho Hypothesis of background information
fair coin - P(EHo) Probability measure
54Notation for Specificity
- E Set of events 100 coin flips
- E One instance of E THTTTHHTHH
- D Function that maps a descriptive language D
to the set of possible events E - D mapping of one event 010001101
- Ho Hypothesis of background information
fair coin - P(EHo) Probability measure
- (f, l) Complexity measure
55Notation for Specificity
- E Set of events 100 coin flips
- E One instance of E THTTTHHTHH
- D Function that maps a descriptive language D
to the set of possible events E - D mapping of one event 010001101
- Ho Hypothesis of background information
fair coin - P(EHo) Probability measure
- (f, l) Complexity measure
- I Side information Binary numbers
56Three conditions to decide in favor of Specified
Pattern over Fabricated Pattern
- Condition 1
- I is conditionally independent of E given Ho
57Three conditions to decide in favor of Specified
Pattern over Fabricated Pattern
- Condition 1
- I is conditionally independent of E given Ho
- Given all possible events E E and given Ho,
what is P(EHo)?
58Three conditions to decide in favor of Specified
Pattern over Fabricated Pattern
- Condition 1
- I is conditionally independent of E given Ho
- Given all possible events E E and given Ho,
what is P(EHo)? - Given all possible events E E and given Ho and
I, what is P(EHo, I)?
59Three conditions to decide in favor of Specified
Pattern over Fabricated Pattern
- Condition 2
- I is sufficient to construct D without
knowledge of E
60Three conditions to decide in favor of Specified
Pattern over Fabricated Pattern
- Condition 2
- I is sufficient to construct D without
knowledge of E - Knowledge of I (binary numbers) gives D
(0100011011) ie. no knowledge of E is necessary
to construct D
61Three conditions to decide in favor of Specified
Pattern over Fabricated Pattern
- Condition 3
- E can be reconstructed from D
- Knowledge of I (binary numbers) gives D
(0100011011) which gives E (THTTTHHTHH) which
was the event that occurred
62Three conditions to decide in favor of Specified
Pattern over Fabricated Pattern
- Condition 1 Conditional Independence
- P(EHo, I) P(EHo)
- Condition 2 Bounded Complexity
- f(DI) lt l
- Condition 3 D delimits E
63SpecificityE, E, D, D, , Ho, P, I, F(f, l)
64SpecificityE, E, D, D, , Ho, P, I, F(f, l)
- Definition Given an event E E, a pattern D
D, whose correspondence maps D into E, a
chance hypothesis Ho, a probability measure P,
where P(Ho) estimates the probability of events
in E given Ho, side information I, and a bounded
complexity measure F (f,l) where f(I)
estimates the difficulty of formulating patterns
in D given I, and l fixes the level of complexity
at which formulating such patterns is feasible,
we say that D is a specification of E relative to
Ho, P, I, and F if and only if the following
conditions are satisfied - Condition 1 P(EHo) P(EHo, I)
- Condition 2 f(DI) lt l
- Condition 3 D delimits E
65Rationale for three Conditions
66Rationale for three conditions
67Rationale for three conditions
- An event occurs
- Probability of event is small
68Rationale for three conditions
- An event occurs
- Probability of event is small
- Assume chance hypothesis is operating to produce E
69Rationale for three conditions
- An event occurs
- Probability of event is small
- Assume chance hypothesis is operating to produce
E - Therefore, I should not give any information
about occurrence of E E (Condition 1)
70Rationale for three conditions
- An event occurs
- Probability of event is small
- Assume chance hypothesis is operating to produce
E - Therefore, I should not give any information
about occurrence of E E (Condition 1) - But, I does give information about D D
(Condition 2)
71Rationale for three conditions
- An event occurs
- Probability of event is small
- Assume chance hypothesis is operating to produce
E - Therefore, I should not give any information
about occurrence of E E (Condition 1) - But, I does give information about D D
(Condition 2) - D gives E (Condition 3)
72Rationale for three conditions
- An event occurs
- Probability of event is small
- Assume chance hypothesis is operating to produce
E - Therefore, I should not give any information
about occurrence of E E (Condition 1) - But, I does give information about D D
(Condition 2) - D gives E (Condition 3)
- Contradiction
73(No Transcript)
74Application
- P(Life occurring on Earth)
75Application
- P(Life occurring on Earth)
- P(Life occurring somewhere in universe)
76Application
- P(Life occurring on Earth)
- P(Life occurring somewhere in universe)
- P(One irreducible biological system occurring
somewhere in universe at some time within the
life of the universe) - Think of eyespot, for example
77Assumptions
- All amino acids available to build proteins
- Ignoring optical isomerization
78Assumptions
- All amino acids available to build proteins
- Ignoring optical isomerization
- Mechanism for building proteins out of amino
acids available
79Assumptions
- All amino acids available to build proteins
- Ignoring optical isomerization
- Mechanism for building proteins out of amino
acids available - Amino acids are completely interchangeable within
each of four types - 8 neutral and hydrophobic
- 7 neutral and hydrophilic
- 3 basic
- 2 acidic
80Assumptions
- Ten proteins required for irreducible system
- Usually requires minimum of 20 - 50
81Assumptions
- Ten proteins required for irreducible system
- Usually requires minimum of 20 - 50
- Protein length of 100 amino acids
- Median length of proteins is 200 400
82Number of Possible Ways to Make 10 Proteins (100
a.a./protein) Using 4 Types of Amino Acids
83Number of Possible Ways to Make 10 Proteins (100
a.a./protein) Using 4 Types of Amino Acids
84Number of Possible Ways to Make 10 Proteins (100
a.a./protein) Using 4 Types of Amino Acids
- (4100)10
- 10602
- P(One irreducible biological system occurring
once) 10-602
85Comment on Small Probability
- When estimating P(EHo), need to consider all
available resources for E to occur
86Available Resources
- P(One irreducible biological system occurring
once) 10-602
87Available Resources
- P(One irreducible biological system occurring
once) 10-602 - What about available resources?
- 1080 elementary particles in universe
88Available Resources
- P(One irreducible biological system occurring
once) 10-602 - What about available resources?
- 1080 elementary particles in universe
- Particles can transition between states at most
1045/sec
89Available Resources
- P(One irreducible biological system occurring
once) 10-602 - What about available resources?
- 1080 elementary particles in universe
- Particles can transition between states at most
1045/sec - Universe is at least 1025 sec old
90Available Resources
- P(One irreducible biological system occurring
once) 10-602 - What about available resources?
- 1080 elementary particles in universe
- Particles can transition between states at most
1045/sec - Universe is at least 1025 sec old
- 1080 1045 1025 10150 chances for an
irreducible biological system to have occurred
91Available Resources
- P(At least one irreducible biological system
occurring somewhere in universe at some time
within the life of the universe)
92Available Resources
- P(At least one irreducible biological system
occurring somewhere in universe at some time
within the life of the universe) - 10-452
93Design Inference
- Small probability
- P(EH,W) 10-452
94Design Inference
- Small probability
- P(EH,W) 10-452
- Is it specified?
95Design Inference
- Small probability
- P(EH,W) 10-452
- Is it specified?
- Irreducible biological systems were designed and
did not happen by chance
96(No Transcript)
97Conclusions
- Most biologists dont understand the mathematics
98Conclusions
- Most biologists dont understand the mathematics
- Most mathematicians dont understand the biology
99Conclusions
- Most biologists dont understand the mathematics
- Most mathematicians dont understand the biology
- Intelligent Design is misunderstood by scientists
100Conclusions
- Most biologists dont understand the mathematics
- Most mathematicians dont understand the biology
- Intelligent Design is misunderstood by scientists
- Evolution is misunderstood by non-scientists
101Conclusions
- Intelligent design is like writing Shakespeares
Romeo and Juliet
102Conclusions
- Intelligent design is like writing Shakespeares
Romeo and Juliet - Evolution is like editing Shakespeares Romeo and
Juliet
103Acknowledgements
- The Design Inference, by William Dembski,
Cambridge University Press, 1998. - Intelligent Design, by William Dembski,
Intervarsity Press, 1999. - Light Antennas in Phototactic Algae, by Kenneth
Foster and Robert Smyth, Microbiological Reviews,
Dec. 1980, pp. 572-630. - Comprehensive Identification of Cell
Cycle-regulated Genes of the Yeast Saccharomyces
cerevisiae by Microarray Hybridization, by
Spellman, et. al., Molecular Biology of the Cell,
Dec. 1998, pp. 3273-3297.
104(No Transcript)
105Fred Hoyle
- Opposed Big Bang Theory and a finite age of the
universe because they were offensive to basic
logical consistency - The trouble is that there are more than 2000
different independent enzymes necessary to life,
and the chances of getting them all produced by a
random trial is less than 10-40,000. This minute
probability of obtaining all the enzymes, only
once each, could not be faced even if the entire
universe consisted of organic soup
106Cell Cycle in Yeast
http//www2.hawaii.edu/alices_v/life.html
107Cell Cycle in Yeast
Cln3p-Cdc28p
108Cell Cycle in Yeast
Cln3p-Cdc28p
109Cell Cycle in Yeast
Cln3p-Cdc28p
MBF SBF
110Cell Cycle in Yeast
Cln3p-Cdc28p
MBF SBF
200 genes
111Cell Cycle in Yeast
Cln3p-Cdc28p
MBF SBF
200 genes
Budding And DNA Synthesis
112Cell Cycle in Yeast
Cln3p-Cdc28p
MBF SBF
200 genes
Budding And DNA Synthesis
Clb2p-Cdc28
113Cell Cycle in Yeast
Cln3p-Cdc28p
MBF SBF
200 genes
Budding And DNA Synthesis
Clb2p-Cdc28
114Cell Cycle in Yeast
Cln3p-Cdc28p
MBF SBF
200 genes
Budding And DNA Synthesis
Clb2p-Cdc28
115Cell Cycle in Yeast
MCM1 SFF
Clb2p-Cdc28
116Cell Cycle in Yeast
MCM1 SFF
Clb2p-Cdc28
117Cell Cycle in Yeast
Swi5p 50 genes
MCM1 SFF
Clb2p-Cdc28
118Cell Cycle in Yeast
SIC1 Other genes
Swi5p 50 genes
MCM1 SFF
Clb2p-Cdc28
119Cell Cycle in Yeast
SIC1 Other genes
Swi5p 50 genes
MCM1 SFF
Clb2p-Cdc28
120Cell Cycle in Yeast
Cln3p-Cdc28p
121http//genome-www.stanford.edu/cellcycle/figures/f
igure1A.html