Complexity of Life and Intelligent Design By Chance or by Design Betsy Siewert Sept' 16, 2006

1
Complexity 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

3
Eyespot 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
4
4 problems to solve in order to swim toward the
light
5
4 problems to solve in order to swim toward the
light

• Sensitivity to detect light
• Light vs. no light

6
4 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

7
4 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

8
4 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

9
Eyespot of Chlamydomonas
http//web.syr.edu/7Esrsangia/research.html
http//web.syr.edu/srsangia/projects.html
10
(No Transcript)
11
Decision Tree Using Design Inference
High Probability?
Intermediate Probability?
Small Probability? Specified?
12
Decision Tree Using Design Inference
High Probability?
Necessity
yes
Intermediate Probability?
Small Probability? Specified?
13
Decision Tree Using Design Inference
High Probability?
Necessity
yes
no
Intermediate Probability?
Small Probability? Specified?
14
Decision Tree Using Design Inference
High Probability?
Necessity
yes
no
Intermediate Probability?
Chance
yes
Small Probability? Specified?
15
Decision Tree Using Design Inference
High Probability?
Necessity
yes
no
Intermediate Probability?
Chance
yes
no
Small Probability? Specified?
16
Decision Tree Using Design Inference
High Probability?
Necessity
yes
no
Intermediate Probability?
Chance
yes
no
Small Probability? Specified?
Design
yes
17
Decision Tree Using Design Inference
18
Is this Specified?

• HHTTTTHHTHTHHTTTHHTHHHHHHHTHTTTHHTTTHHTHHTTHHHTHH
  HTTTHHTTHTTTTHTHHHHTHHHTHHTTHHHHHTHTTHTHTTHTHTHTHH
  HTTHHHHHH

19
Is this Specified?

• HTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTH
  THTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTHTH
  THTHTHTHT

20
Example of Using Decision Tree
21
Assumptions

• 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

22
How many possible combinations are there?
23
How many possible combinations are there?

• 100100100100100
• 1005

24
How many possible combinations are there?

• 100100100100100
• 1005
• 10,000,000,000
• (10 billion)

25
Decision Tree Using Design Inference
26
(No Transcript)
27
Probability

• 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)

28
If Background information (Ho) changes, then
probability (P) might change

• E Event that John will go to a party on Friday

29
If 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

30
If 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)?

31
If 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

32
If 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)

33
If 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)?

34
If 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

35
If 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)

36
Complexity

• 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
37
Complexity and Probability

• Complexity and Probability are inversely related
• Probability Complexity
• P(EHo) f (QR)
• 0,1 0, 8)

38
Complexity and Probability

• Shannon information is a type of measure of
  complexity
• I -log2 (p)

39
Complexity and Probability

• Shannon information is a type of measure of
  complexity
• I -log2 (p)
• Bank Safe
• f (Q) -log2 (10-10) 33

40
Tractability Bounds for Complexity

• f(QR) lt l Q is tractable
• f(QR) gt m Q is intractable
• Notation (f, l)

41
(No Transcript)
42
Specificity

• Pattern
• Specified pattern
• Fabricated pattern
• Example

43
Specificity

• Pattern
• Specified pattern
• Fabricated pattern
• Example

http//www.fotosearch.com
44
Example

• THTTTHHTHHTTTTTHTHTTHHHTTHTHHHTHHHTTTTTTTHTTHTTTHH
  THTTTHTHTHHTTHHHHTTTHTTHHTHTHTHHHHTTHHTHHHHTHHHHTT
  

45
Example

• THTTTHHTHHTTTTTHTHTTHHHTTHTHHHTHHHTTTTTTTHTTHTTTHH
  THTTTHTHTHHTTHHHHTTTHTTHHTHTHTHHHHTTHHTHHHHTHHHHTT
  
• 01000110110000010100111001011101110000000100100011
  01000101011001111000100110101011110011011110111100
  

46
Example

• 01000110110000010100111001011101110000000100100011
  01000101011001111000100110101011110011011110111100
  

47
Example

• 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

48
Notation for Specificity

• E Set of events All possible results from 100
  coin flips

49
Notation for Specificity

• E Set of events 100 coin flips
• E One instance of E THTTTHHTHH

50
Notation 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

51
Notation 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

52
Notation 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

53
Notation 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

54
Notation 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

55
Notation 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

56
Three conditions to decide in favor of Specified
Pattern over Fabricated Pattern

• Condition 1
• I is conditionally independent of E given Ho

57
Three 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)?

58
Three 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)?

59
Three conditions to decide in favor of Specified
Pattern over Fabricated Pattern

• Condition 2
• I is sufficient to construct D without
  knowledge of E

60
Three 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

61
Three 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

62
Three 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

63
SpecificityE, E, D, D, , Ho, P, I, F(f, l)
64
SpecificityE, 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

65
Rationale for three Conditions
66
Rationale for three conditions

• An event occurs

67
Rationale for three conditions

• An event occurs
• Probability of event is small

68
Rationale for three conditions

• An event occurs
• Probability of event is small
• Assume chance hypothesis is operating to produce E

69
Rationale 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)

70
Rationale 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)

71
Rationale 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)

72
Rationale 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)
74
Application

• P(Life occurring on Earth)

75
Application

• P(Life occurring on Earth)
• P(Life occurring somewhere in universe)

76
Application

• 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

77
Assumptions

• All amino acids available to build proteins
• Ignoring optical isomerization

78
Assumptions

• All amino acids available to build proteins
• Ignoring optical isomerization
• Mechanism for building proteins out of amino
  acids available

79
Assumptions

• 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

80
Assumptions

• Ten proteins required for irreducible system
• Usually requires minimum of 20 - 50

81
Assumptions

• 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

82
Number of Possible Ways to Make 10 Proteins (100
a.a./protein) Using 4 Types of Amino Acids

• (4100)10

83
Number of Possible Ways to Make 10 Proteins (100
a.a./protein) Using 4 Types of Amino Acids

• (4100)10
• 10602

84
Number 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

85
Comment on Small Probability

• When estimating P(EHo), need to consider all
  available resources for E to occur

86
Available Resources

• P(One irreducible biological system occurring
  once) 10-602

87
Available Resources

• P(One irreducible biological system occurring
  once) 10-602
• What about available resources?
• 1080 elementary particles in universe

88
Available 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

89
Available 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

90
Available 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

91
Available Resources

• P(At least one irreducible biological system
  occurring somewhere in universe at some time
  within the life of the universe)

92
Available Resources

• P(At least one irreducible biological system
  occurring somewhere in universe at some time
  within the life of the universe)
• 10-452

93
Design Inference

• Small probability
• P(EH,W) 10-452

94
Design Inference

• Small probability
• P(EH,W) 10-452
• Is it specified?

95
Design 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)
97
Conclusions

• Most biologists dont understand the mathematics

98
Conclusions

• Most biologists dont understand the mathematics
• Most mathematicians dont understand the biology

99
Conclusions

• Most biologists dont understand the mathematics
• Most mathematicians dont understand the biology
• Intelligent Design is misunderstood by scientists

100
Conclusions

• Most biologists dont understand the mathematics
• Most mathematicians dont understand the biology
• Intelligent Design is misunderstood by scientists
• Evolution is misunderstood by non-scientists

101
Conclusions

• Intelligent design is like writing Shakespeares
  Romeo and Juliet

102
Conclusions

• Intelligent design is like writing Shakespeares
  Romeo and Juliet
• Evolution is like editing Shakespeares Romeo and
  Juliet

103
Acknowledgements

• 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)
105
Fred 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

106
Cell Cycle in Yeast
http//www2.hawaii.edu/alices_v/life.html
107
Cell Cycle in Yeast
Cln3p-Cdc28p
108
Cell Cycle in Yeast
Cln3p-Cdc28p
109
Cell Cycle in Yeast
Cln3p-Cdc28p
MBF SBF
110
Cell Cycle in Yeast
Cln3p-Cdc28p
MBF SBF
200 genes
111
Cell Cycle in Yeast
Cln3p-Cdc28p
MBF SBF
200 genes
Budding And DNA Synthesis
112
Cell Cycle in Yeast
Cln3p-Cdc28p
MBF SBF
200 genes
Budding And DNA Synthesis
Clb2p-Cdc28
113
Cell Cycle in Yeast
Cln3p-Cdc28p
MBF SBF
200 genes
Budding And DNA Synthesis
Clb2p-Cdc28
114
Cell Cycle in Yeast
Cln3p-Cdc28p
MBF SBF
200 genes
Budding And DNA Synthesis
Clb2p-Cdc28
115
Cell Cycle in Yeast
MCM1 SFF
Clb2p-Cdc28
116
Cell Cycle in Yeast
MCM1 SFF
Clb2p-Cdc28
117
Cell Cycle in Yeast
Swi5p 50 genes
MCM1 SFF
Clb2p-Cdc28
118
Cell Cycle in Yeast
SIC1 Other genes
Swi5p 50 genes
MCM1 SFF
Clb2p-Cdc28
119
Cell Cycle in Yeast
SIC1 Other genes
Swi5p 50 genes
MCM1 SFF
Clb2p-Cdc28
120
Cell Cycle in Yeast
Cln3p-Cdc28p
121
http//genome-www.stanford.edu/cellcycle/figures/f
igure1A.html