DNA Computing and Molecular Programming

DNA Computing and Molecular Programming
Byoung-Tak Zhang Biointelligence Laboratory School of Computer Science and Engineering Seoul National University Seoul , Korea I will talk about evolving DNA-encoded genetic programs in a test tube. We evaluate the potentials of this approach by solving a medical diagnosis problem on a simulated DNA computer. The individual genetic program represents a decision list of variable length and the whole population takes part in making probabilistic decisions.

Eigen’s Evolution Machine
© 2005, SNU Biointelligence Lab,

DNA Computing

Motivation: In Vitro Evolution as EC Technology
Each DNA molecule represents an individual (or a genetic program) at nanoscale A huge population of up to Avogadro number (6 x 1023) of molecules Molecular recognition by chemistry Exponential self-replication by PCR Massively parallel variation-selection operators Ultra-low energy consumption Evolvable “wet” “molecular” hardware © 2005, SNU Biointelligence Lab,

Why Try Molecular EC? 6.022  1023 molecules / mole Massively Parallel Search Desktop: 109 operations / sec Supercomputer: 1012 operations / sec 1 mmol of DNA: 1026 reactions Favorable Energetics: Gibbs Free Energy 1 J for 2  1019 operations Storage Capacity: 1 bit per cubic nanometer The fastest supercomputer vs. DNA computer 106 op/sec vs op/sec 109 op/J vs op/J (in ligation step) 1bit per 1012 nm3 vs. 1 bit per 1 nm3 (video tape vs. molecules) © 2005, SNU Biointelligence Lab,

Molecular Programming (MP): Evolving Genetic Programs in a Test Tube
Theory Bayesian evolution [Zhang, CEC-99; Zhang, Handbook-2003] Model Probabilistic library model [Zhang, DNA-04 & DNA-05] Algorithm Molecular algorithms [Zhang, GP-98] Representation Decision lists Operators Molecular operators for variation and selection © 2005, SNU Biointelligence Lab,

DNA Molecular Computing
Nanostructure Molecular recognition Self-assembly Self-replication Heat Cool Polymer Repeat © 2007, SNU Biointelligence Lab,

Encoding a Hypernetwork with DNA
z1 : z2 : z3 : z4 : b) x1 x2 x3 x4 x5 y 1 where z1 : (x1=0, x2=1, x3=0, y=1) z2 : (x1=0, x2=0, x3=1, x4=0, x5=0, y=0) z3 : (x2=1, x4=1, y=1) z4 : (x2=1, x3=0, x4=1, y=0) a) AAAACCAATTGGAAGGCCATGCGG AAAACCAATTCCAAGGGGCCTTCCCCAACCATGCCC AATTGGCCTTGGATGCGG AATTGGAAGGCCCCTTGGATGCCC GG AAAA AATT AAGG CCTT CCAA ATGC CC Collection of (labeled) hyperedges Library of DNA molecules corresponding to (a) For example, a program x sub one equals one and x sub three equals one and x sub five equals one and y equals one in the form of decision lists or its DNA encoding denotes a decision rule saying diagnose the DNA sample as positive for disease y if contains all the three markers x sub one, x sub three and x sub five. © 2007, SNU Biointelligence Lab,

Initial Library L0 (x1=0, x2=1, x3=1, y=1) (x2=1, y=0) AAAACCAATTGGAATTGGATGCGG AATTGGATGCCC (x1=0, x2=0, x3=1, y=0) (x2=1, x3=0, y=0) AAAACCAATTCCAAGGGGATGCCC (x1=0, y=0) AAAACCATGCCC AATTGGAAGGCCATGCCC (x2=1, x3=1, y=1) AATTGGCCTTGGATGCGG (x1=0, y=1) AAAACCATGCGG (x2=0, y=0) AATTCCATGCCC (x2=0, y=1) AATTCCATGCGG … (x1=0, x2=0, y=0) AAAACCAATTCCATGCCC (x1=0, x2=0, y=1) (x1=0, x2=1, y=0) AAAACCAATTCCATGCGG AAAACCAATTGGATGCCC AAAACCAATTCCATGCGG AAAACCAATTGGATGCCC AAAACCAATTCCATGCGG AAAACCAATTGGATGCCC (x1=0, x2=1, y=1) … (x1=0, x2=0, x3=0, y=0) AAAACCAATTCCAAGGCCATGCCC AAAACCAATTGGATGCGG AAAACCAATTGGATGCGG AAAACCAATTGGATGCGG (x1=0, x2=0, x3=0, y=1) (x1=0, x2=0, x3=1, y=0) AAAACCAATTCCAAGGCCATGCGG AAAACCAATTCCAAGGGGATGCCC AAAACCAATTCCAAGGCCATGCGG AAAACCAATTCCAAGGGGATGCCC AAAACCAATTCCAAGGCCATGCGG AAAACCAATTCCAAGGGGATGCCC (x1=0, x2=0, x3=1, y=1) (x1=0, x2=1, x3=0, y=0) AAAACCAATTCCAAGGGGATGCGG AAAACCAATTGGAAGGCCATGCCC AAAACCAATTCCAAGGGGATGCGG AAAACCAATTGGAAGGCCATGCCC AAAACCAATTCCAAGGGGATGCGG AAAACCAATTGGAAGGCCATGCCC (x1=0, x2=1, x3=0, y=1) AAAACCAATTGGAAGGCCATGCGG … x1 x2 x3 y 1 where AAGG AATT AAAA ATGC CC GG © 2005, SNU Biointelligence Lab,

+ Hybridization Library Example 1 (x1=0, x2=1, x3=1, y=1) (x1=0, x2=1, x3=1, y=1) (x1=0, x2=1, x3=0, y=0) AAAACCAATTGGAATTGGATGCGG AAAACCAATTGGAATTGGATGCGG TTTTGG TTAACC TTCCGG TTTTGG TTAACC TACGGG (x1=0, x2=0, x3=1, y=0) (x1=0, x2=0, x3=1, y=0) TTTTGG TTAACC TTCCGG TACGGG AAAACCAATTCCAAGGGGATGCCC AAAACCAATTCCAAGGGGATGCCC TTTTGG GGTTGG (x2=1, x3=1, y=1) TTTTGG TTAACC TTCCGG TACGGG AATTGGCCTTGGATGCGG (x2=1, x3=1, y=1) AATTGGCCTTGGATGCGG (x2=1, x3=0, y=0) TTAACC AATTGGAAGGCCATGCCC (x2=1, x3=0, y=0) AATTGGAAGGCCATGCCC TTAACC TTCCGG GGTTGG (x2=1, y=0) AATTGGATGCCC (x2=1, y=0) Amplify AATTGGATGCCC TTAACC GGTTGG © 2005, SNU Biointelligence Lab,

Updated Library L1 (x1=0, x2=1, x3=1, y=1) (x2=1, y=0) AAAACCAATTGGAATTGGATGCGG AATTGGATGCCC AATTGGATGCCC (x1=0, x2=0, x3=1, y=0) (x2=1, x3=0, y=0) AAAACCAATTCCAAGGGGATGCCC (x1=0, y=0) AAAACCATGCCC AATTGGAAGGCCATGCCC (x2=1, x3=1, y=1) AATTGGAAGGCCATGCCC AATTGGCCTTGGATGCGG (x1=0, y=1) AAAACCATGCGG (x2=0, y=0) AATTCCATGCCC (x2=0, y=1) AATTCCATGCGG … (x1=0, x2=0, y=0) AAAACCAATTCCATGCCC (x1=0, x2=0, y=1) (x1=0, x2=1, y=0) AAAACCAATTCCATGCGG AAAACCAATTGGATGCCC AAAACCAATTCCATGCGG AAAACCAATTGGATGCCC AAAACCAATTCCATGCGG AAAACCAATTGGATGCCC (x1=0, x2=1, y=1) … (x1=0, x2=0, x3=0, y=0) AAAACCAATTCCAAGGCCATGCCC AAAACCAATTGGATGCGG AAAACCAATTGGATGCGG AAAACCAATTGGATGCGG (x1=0, x2=0, x3=0, y=1) (x1=0, x2=0, x3=1, y=0) AAAACCAATTCCAAGGCCATGCGG AAAACCAATTCCAAGGGGATGCCC AAAACCAATTCCAAGGCCATGCGG AAAACCAATTCCAAGGGGATGCCC AAAACCAATTCCAAGGCCATGCGG AAAACCAATTCCAAGGGGATGCCC (x1=0, x2=0, x3=1, y=1) (x1=0, x2=1, x3=0, y=0) AAAACCAATTCCAAGGGGATGCGG AAAACCAATTGGAAGGCCATGCCC AAAACCAATTCCAAGGGGATGCGG AAAACCAATTGGAAGGCCATGCCC AAAACCAATTCCAAGGGGATGCGG AAAACCAATTGGAAGGCCATGCCC (x1=0, x2=1, x3=0, y=1) AAAACCAATTGGAAGGCCATGCGG … x1 x2 x3 y 1 where AAGG AATT AAAA ATGC CC GG © 2005, SNU Biointelligence Lab,

+ Hybridization Library Example 2 (x1=0, x2=1, x3=1, y=1) (x1=0, x2=1, x3=1, y=1) (x1=0, x2=1, x3=1, y=1) AAAACCAATTGGAATTGGATGCGG AAAACCAATTGGAATTGGATGCGG TTTTGG TTAACC TTCCCC TTTTGG TTAACC TTCCCC TACGCC TACGCC (x1=0, x2=0, x3=1, y=0) (x1=0, x2=0, x3=1, y=0) TTTTGG TTAACC TTCCCC TACGCC AAAACCAATTCCAAGGGGATGCCC AAAACCAATTCCAAGGGGATGCCC TTTTGG TTCCCC Amplify (x2=1, x3=1, y=1) TTTTGG TTAACC TTCCCC TACGCC (x2=1, x3=1, y=1) AATTGGCCTTGGATGCGG AATTGGCCTTGGATGCGG (x2=1, x3=0, y=0) TTAACC TTCCCC TACGCC AATTGGAAGGCCATGCCC (x2=1, x3=0, y=0) (x2=1, x3=0, y=0) AATTGGAAGGCCATGCCC AATTGGAAGGCCATGCCC TTAACC AATTGGAAGGCCATGCCC (x2=1, y=0) AATTGGATGCCC TTAACC AATTGGATGCCC (x2=1, y=0) (x2=1, y=0) AATTGGATGCCC TTAACC AATTGGATGCCC TTAACC © 2005, SNU Biointelligence Lab,

Updated Library L2 (x1=0, x2=1, x3=1, y=1) (x2=1, y=0) AAAACCAATTGGAATTGGATGCGG AATTGGATGCCC AAAACCAATTGGAATTGGATGCGG AATTGGATGCCC (x1=0, x2=0, x3=1, y=0) (x2=1, x3=0, y=0) AAAACCAATTCCAAGGGGATGCCC (x1=0, y=0) AAAACCATGCCC AATTGGAAGGCCATGCCC (x2=1, x3=1, y=1) AATTGGAAGGCCATGCCC AATTGGCCTTGGATGCGG AATTGGCCTTGGATGCGG (x1=0, y=1) AAAACCATGCGG (x2=0, y=0) AATTCCATGCCC (x2=0, y=1) AATTCCATGCGG … (x1=0, x2=0, y=0) AAAACCAATTCCATGCCC (x1=0, x2=0, y=1) (x1=0, x2=1, y=0) AAAACCAATTCCATGCGG AAAACCAATTGGATGCCC AAAACCAATTCCATGCGG AAAACCAATTGGATGCCC AAAACCAATTCCATGCGG AAAACCAATTGGATGCCC (x1=0, x2=1, y=1) … (x1=0, x2=0, x3=0, y=0) AAAACCAATTCCAAGGCCATGCCC AAAACCAATTGGATGCGG AAAACCAATTGGATGCGG AAAACCAATTGGATGCGG (x1=0, x2=0, x3=0, y=1) (x1=0, x2=0, x3=1, y=0) AAAACCAATTCCAAGGCCATGCGG AAAACCAATTCCAAGGGGATGCCC AAAACCAATTCCAAGGCCATGCGG AAAACCAATTCCAAGGGGATGCCC AAAACCAATTCCAAGGCCATGCGG AAAACCAATTCCAAGGGGATGCCC (x1=0, x2=0, x3=1, y=1) (x1=0, x2=1, x3=0, y=0) AAAACCAATTCCAAGGGGATGCGG AAAACCAATTGGAAGGCCATGCCC AAAACCAATTCCAAGGGGATGCGG AAAACCAATTGGAAGGCCATGCCC AAAACCAATTCCAAGGGGATGCGG AAAACCAATTGGAAGGCCATGCCC (x1=0, x2=1, x3=0, y=1) AAAACCAATTGGAAGGCCATGCGG … x1 x2 x3 y 1 where AAGG AATT AAAA ATGC CC GG © 2005, SNU Biointelligence Lab,

+ Hybridization Library Query (x1=0, x2=1, x3=1, y=1) (x1=0, x2=1, x3=1, y=1) (x1=1, x2=1, x3=0) AAAACCAATTGGAATTGGATGCGG AAAACCAATTGGAATTGGATGCGG TTAACC TTTTCC TTAACC TTCCGG AAAACCAATTGGAATTGGATGCGG AAAACCAATTGGAATTGGATGCGG TTAACC (x1=0, x2=0, x3=1, y=0) TTTTCC TTAACC TTCCGG (x1=0, x2=0, x3=1, y=0) AAAACCAATTCCAAGGGGATGCCC AAAACCAATTCCAAGGGGATGCCC TTTTCC TTAACC TTCCGG (x2=1, x3=1, y=1) (x2=1, x3=1, y=1) AATTGGCCTTGGATGCGG AATTGGCCTTGGATGCGG Predict the class TTAACC AATTGGCCTTGGATGCGG AATTGGCCTTGGATGCGG TTAACC (x2=1, x3=0, y=0) (x2=1, x3=0, y=0) AATTGGAAGGCCATGCCC AATTGGAAGGCCATGCCC TTAACC TTCCGG AATTGGAAGGCCATGCCC AATTGGAAGGCCATGCCC (x2=1, y=0) TTAACC TTCCGG (x2=1, y=0) AATTGGATGCCC AATTGGATGCCC TTAACC AATTGGATGCCC AATTGGATGCCC TTAACC Majority voting © 2005, SNU Biointelligence Lab,

Molecular Information Processing
MP4.avi © 2007, SNU Biointelligence Lab,

Molecular Self-Assembly of Hypernetworks
xi xj y Molecular Encoding Hypernetwork Representation X1 X2 X8 X3 X7 X4 X6 X5 © 2007, SNU Biointelligence Lab,

Learning the Hypernetwork (by Molecular Evolution)
Next generation Library of combinatorial molecules Library Example + The aim is to build a decision making system f that outputs label Select the library elements matching the example Amplify the matched library elements by PCR Hybridize [Zhang, DNA11] © 2007, SNU Biointelligence Lab,

The Theory of Bayesian Evolution Evolution as a Bayesian inference process Evolutionary computation (EC) is viewed as an iterative process of generating the individuals of ever higher posterior probabilities from the priors and the observed data. generation 0 generation g P(A |D) P(A |D) ... P0(Ai) Pg(Ai |D) Pg(Ai) [Zhang, CEC-99] © 2007, SNU Biointelligence Lab,

Evolutionary Learning Algorithm for Hypernetwork Classifiers
1. Let the hypernetwork H represent the current distribution P(X,Y). 2. Get a training example (x,y). 3. Classify x using H as follows 3.1 Extract all molecules matching x into M. 3.2 From M separate the molecules into classes: Extract the molecules with label Y=0 into M0 Extract the molecules with label Y=1 into M1 3.3 Compute y*=argmaxY{0,1}| MY |/|M| 4. Update H If y*=y, then Hn ← Hn-1+{c(u, v)} for u=x and v=y for (u, v) Hn-1, If y*≠y, then Hn ← Hn-1{c(u, v)} for u=x and v ≠ y for (u, v) Hn-1 5.Goto step 2 if not terminated. © 2007, SNU Biointelligence Lab,

Molecular Operators

Molecular Operators Variation Ligation Restriction Mutation (PCR) Selection Gel electrophoresis Affinity separation (beads) Capillary electrophoresis Amplification Polymerase chain reaction (PCR) Rolling circle amplification (RCA) © 2005, SNU Biointelligence Lab,

Variation: Hybridization & Ligation
base-pairing between two complementary single-strand molecules to form a double stranded DNA molecule Ligation Joining DNA molecules together Usually used for candidate solution generation. Hybridization Ligation © 2005, SNU Biointelligence Lab,

Variation: Restriction
Cut the specific DNA site. Solution detection or filtering step A A G C T T T T C G A A OH 3’ A 5’ P A C G T T EcoRI T T C G A A P 5’ 3’ OH © 2005, SNU Biointelligence Lab,

Selection: Gel Electrophoresis
Detection desired solutions. Separate solution molecules by length © 2005, SNU Biointelligence Lab,

Amplification: PCR Polymerase chain reaction Amplifies (produces identical copies of) selected dsDNA molecules. Make 2n copies (n : number of iteration) Used to filter solutions or detection. © 2005, SNU Biointelligence Lab,

DNA Computing and Molecular Programming

Similar presentations

Presentation on theme: "DNA Computing and Molecular Programming"— Presentation transcript:

Similar presentations

About project

Feedback

Log in

Auth with social network:

DNA Computing and Molecular Programming

Similar presentations

Presentation on theme: "DNA Computing and Molecular Programming"— Presentation transcript:

Similar presentations

About project

Feedback