Outline of NACST/Sim 2003.4.8 신수용.

Presentation on theme: "Outline of NACST/Sim 2003.4.8 신수용."— Presentation transcript:

1 Outline of NACST/Sim 신수용

2 © 2003, SNU BioIntelligence Lab, http://bi.snu.ac.kr/
Basic concept Implementing Hybridization first. Then, ligation, PCR, and Bead seperation Can be implemented easily based on hybridization concept. 그외.. Gel electrophoresis is easy. Bead seperation은 아직 고려하지 않음 © 2003, SNU BioIntelligence Lab,

3 © 2003, SNU BioIntelligence Lab, http://bi.snu.ac.kr/
Hybridization model Probabilistic model. Sequence Collision Probability Tube에서 sequence들이 어떻게 분포하고 있으며, 어떤 sequence들이 서로 만날 것인가에 대한 모델 Sequence Hybridization Probability 서로 만난 sequence들의 결합가능성 판단 © 2003, SNU BioIntelligence Lab,

4 © 2003, SNU BioIntelligence Lab, http://bi.snu.ac.kr/
Collision model Based on Artificial Chemistry Artificial Chemistries: A Review, P. Dittrich, J. Ziegler, and W. Banzhaf Multi-Agent Systems inspired by Artificial Chemistries: A Case Study in Automated Theorem Proving, J. Bush and W. Banzhaf © 2003, SNU BioIntelligence Lab,

5 © 2003, SNU BioIntelligence Lab, http://bi.snu.ac.kr/
Collision model (AC) 현재의 DNAC를 위해서는 초기에 입력은 전부 주어지며, 출력도 없는 단순한 모델로 충분. Sequence들은 uniform distribution에 근거하여 확률적으로 선택됨 Hybridization이 되어 긴 sequence가 새로 생기면 추가적인 input으로 가정 © 2003, SNU BioIntelligence Lab,

6 © 2003, SNU BioIntelligence Lab, http://bi.snu.ac.kr/
Collision model 새로 생성된 긴 DNA © 2003, SNU BioIntelligence Lab,

7 © 2003, SNU BioIntelligence Lab, http://bi.snu.ac.kr/
Hybridization model 2개의 module로 구성됨 하나는 MOEA의 decision maker로 사용 하나는 virtual test tube로 사용 Internal one base mismatch 고려 특수한 경우의 two base mismatch 고려 Terminal mismatch 고려 Dangling ends 고려 Internal loop 고려 Secondary structure 고려 © 2003, SNU BioIntelligence Lab,

8 Hybridization Model: First module
DNA sequence set이 입력으로 들어가고 2개씩을 순서대로 선택해서 각각 shift를 시켜가면서 결합가능성 검사 주용도: NACST/Seq에서 만든 sequence set들 중에서 가장 적절한 sequences를 선택 in silico experiment를 통해 가장 적절한 sequence를 선택 (vs. in vitro selection) © 2003, SNU BioIntelligence Lab,

9 Hybridization Model: Second module
Based on Sequence Alignment Algorithms Dynamic Programming 좀 더 빠른 알고리즘을 만들 필요 있음 독일에서 만든 알고리즘은 시퀀스 2개 주고 결과를 얻기 위해서는 대략 5-9초 정도의 시간이 필요. DNASIM: 1998, Jens Niehaus 기본 가정이 NN model이 아닌 virtual stack을 사용한 것으로 보임. 지금 현재 개발중. © 2003, SNU BioIntelligence Lab,

10 Hybridization: 결합 결정 기준
단순히 Tm보다 크다고 해서 붙고 작다고 해서 붙지 않는 것은 현실 상황과 차이가 크다고 생각됨. Tm이 sigmoid function이라는 것을 고려하여 확률적으로 결합여부를 결정할 필요 있을 것으로 고려됨. © 2003, SNU BioIntelligence Lab,

11 Hybridization: 결합 결정 기준
Tm은 α = 0.5일 때임 Fraction of total DNA in double-stranded form Current temperature © 2003, SNU BioIntelligence Lab,

12 Nearest-neighbor model for Tm
R : Boltzmann’s constant (1.987 cal/(K mol)) [C]] : total molar strand concentration T : Kelvin [Na+] concentrations different from 1M © 2003, SNU BioIntelligence Lab,

13 Nearest-neighbor model for Tm
Self-complement : [CT]/4  [CT] © 2003, SNU BioIntelligence Lab,

14 Nearest-neighbor data
Thermodynamics and NMR of Internal G.T Mismatches in DNA, H. T. Allawi and J. SantaLucia, Jr., Biochemistry 1997, 36, Nearest Neighbor Thermodynamic Parameters for Internal GA Mismatches in DNA, H. T. Allawi, J. SantaLucia, Jr., Biochemistry 1998, 37, Thermodynamics of internal CT mismatches in DNA, H. T. Allawi and J. SantaLucia, Jr., Nucleic Acids Research, 1998, 26(11): Nearest-Neighbor Thermodynamics of Internal AC Mismatches in DNA: Sequence Dependence and pH Effects, H. T. Allawi and J. SantaLucia, Jr., Biochemistry, 1998, 37, Nearest-Neighbor Thermodynamics and NMR of DNA sequences with Internal AA, CC, GG, and TT Mismatches, N. Peyret, P. A. Senevirante, H. T. Allawi, and J. SantaLucia, Jr., Biochemistry 1999, 38, The Effect of Base Sequence on the Stability of RNA and DNA Single Base Bulges, J. Zhu and R. M. Wartell, Biochemistry 1999, 38, Thermodynamic parameters for DNA sequences with dangling ends, S. Bommarito, N. Peyret, and J. SantaLucia, Jr., Nucleic Acids Research, 2000, 28(9): © 2003, SNU BioIntelligence Lab,

15 NN parameters (Delta S): with internal one base mismatch
© 2003, SNU BioIntelligence Lab,

16 NN parameters (Delta H): with internal one base mismatch
© 2003, SNU BioIntelligence Lab,

17 NN parameters: Dangling ends
© 2003, SNU BioIntelligence Lab,

18 NN parameters: terminal mismatch
Terminal mismatch can be expressed as the sum of the two dangling end motifs Ex) GA = GA G CA C CA 정확한 Data가 SantaLucia Lab에 있으나 private data 위와 같이 할 경우 정확하지는 않으나 오차가 아주 클 것으로 생각되지는 않음 48개의 data에 대해서 8개는 위의 경우가 0.4kcal/mol 이상 stable 29개는 within 0.4kcal/mol 11개는 0.4 kcal/mol 이상으로 unstable © 2003, SNU BioIntelligence Lab,

19 NN parameter: single base bulge
Base bulge가 인접 base 둘 중 하나라도 동일한 경우 동일하지 않은 경우 © 2003, SNU BioIntelligence Lab,

20 © 2003, SNU BioIntelligence Lab, http://bi.snu.ac.kr/
추후 추가 예정 data DNA secondary structure prediction model이 필요 일단은 RNA secondary structure prediction algorithm을 사용할 생각임. Hairpin Internal loop bulge loop pseudoknot Branch loop junction duplex © 2003, SNU BioIntelligence Lab,

21 © 2003, SNU BioIntelligence Lab, http://bi.snu.ac.kr/
For implementation We need to know HOW MANY MOLECULES ARE NECESSARY? HOW MUCH TIME IS NECESSARY? © 2003, SNU BioIntelligence Lab,

22 Hybridization reaction analysis
Simple 3 nodes directed graph DNA strands © 2003, SNU BioIntelligence Lab,

23 Hybridization reaction: kinetics equations
© 2003, SNU BioIntelligence Lab,

24 Hybridization reaction: kinetics equations
Perfect match case Steady-state approximation 초기값 : α Minimum error rate : ε © 2003, SNU BioIntelligence Lab,

25 Initial volume & Minimum computation time
α : 10 nM k : 107 M-1s-1 If EtBr is used (gel electrophoresis dye) : ε is 1.6ⅹ10-7 M © 2003, SNU BioIntelligence Lab,