SB Course

From CSBLwiki

Latest revision as of 04:31, 15 December 2009

Biomodel database

• Biomodel DB@EBI
• Simulating Biomodels in JWS online

Running SBML

• SBML softwares - guide table
• KEGG2SBML - KEGG pathway converter to SBML (written in Perl)

R-project

• R package
• BioConductor - based on R package
  • SBMLR package - Manual
    • getIncidenceMatrix (get incidence/stoichiometry matrix from model)
  • RSBML - quick introduction

Matlab

• SimBiology Toolbox provided by Mathworks

M = getstoichmatrix(modelObj)
[M,objSpecies] = getstoichmatrix(modelObj)
[M,objSpecies,objReactions] = getstoichmatrix(modelObj)

• Systems Biology Toolbox 2 - a third party toolbox <=== this link seemed broken or blocked
• Systems Biology Toolbox - try this! <=== it might be an old stuff
  • Download SBTOOLBOX.zip
• SBMLToolbox by SBML.org
  • Manual

Readings

• Books
  • Stochastic Modelling for Systems Biology by DJ Wilkinson (2006, Chapman & Hall/CRC press)
  • An Introduction to Systems Biology: Design Principles of Biological Circuits by Uri Alon (2006, Chapman & Hall/CRC press)
• Math tutorials
  • Ken Miller's Math Notes - UCSF
• Metabolic Control Analysis (MCA)
  • Tutorial

Term project

2009 Fall

• 지난수업시간에 공지한것처럼 기말 프로젝트로 다음 세가지 옵션중 1개를 선택하여 12/18(금)까지 제출해 주십시오.

1. Matlab을 이용한 SBML 읽기 및 모델링 - Biomodel database (http://www.ebi.ac.uk/biomodels-main/) 에서 테스트를 원하는 모델을 SBML형식으로 다운로드 받아서 Matlab에서 읽고, Default parameter로 시뮬레이션을 합니다 (제출해야할것 - 작동하는 Matlab 코드, 사용한 SBML화일) <=== 현재 the third party toolbox gives errors to handle BIOMODEL SBMLs. You may modify the SBML file (e.g. symbols)
2. R package를 사용한 SBML읽기 - 현재 SBMLR library package가 불완전하여 모든 SBML을 성공적으로 읽지 못합니다. 따라서 KEGG에 있는 sde00010 pathway (glycolysis in Saccharophagus degradans)를 SBMLR이 읽을 수 있는 형태 (e.g. readSBMLR)로 변환하여 R 에서 읽습니다 (제출해야할것 - R 의 SBMLR이 읽을수 있는 sde00010 pathway 화일)
3. Matlab이나 R을 사용하지 하는 경우, KEGG에 있는 sde00010 pathway의 전체 stoichiometric matrix(S)를 MS-Excel화일에 직접 작성합니다. 이때, row는 species, column은 reaction으로 합니다 (제출해야할것 - stoichiometric matrix를 포함한 MS-Excel화일)

• Reconstructing central carbon pathways in target organims
  • Shewanella oneidensis MR-1 KEGG pathway map
    • Lactate metabolism
    • PubMed on flux analysis
  • Saccharophagus degradans 2-40 KEGG pathway map
    • Glucose to Lactate
    • converting KEGG pathway map into 'sbml' by kegg2sbml
    • or using Keggconverter
    • Example
      • glycolysis pathway map - sde0010
      • detailed information (content) - sde0010
      • download pathway map - KEGG pathway_gif ftp - The final GIF version of KEGG PATHWAY (to be removed)
• Example using SBMLR
  • Reference

1. Curto, R., Voit, E. O. and Cascante, M. Analysis of abnormalities in purine metabolism leading to gout and to neurological dysfunctions in man. Biochem.J. 329 (Pt 3), 477-487 (1998a).
2. Curto, R., Voit, E. O., Sorribas, A. and Cascante, M. Validation and steady-state analysis of a power-law model of purine metabolism in man. Biochem.J. 324 (Pt 3), 761-775 (1997).
3. Curto, R., Voit, E. O., Sorribas, A. and Cascante, M. Mathematical models of purine metabolism in man. Math.Biosci. 151, 1-49 (1998b)

• R-script

## installing Bioconductor
source("http://bioconductor.org/biocLite.R")
biocLite()                               # install default packages
biocLite(c("KEGGgraph","SBMLR"))         # install specified packages
## loading packages by 'library' function
library(KEGGgraph)
##
##---- The following example performs a perturbation in PRPP from 5 to 50 uM in Curto et al. 
##
library(SBMLR) 
library(odesolve) 
curto=readSBML(file.path(system.file(package="SBMLR"), "models/curto.xml"))
summary(curto)
names(curto)
out1=simulate(curto,seq(-20,0,1)) 
curto$species$PRPP$ic=50 
out2=simulate(curto,0:70) 
outs=data.frame(rbind(out1,out2))

attach(outs) 
par(mfrow=c(2,1)) 
plot(time,IMP,type="l") 
plot(time,HX,type="l") 
par(mfrow=c(1,1)) 
detach(outs)

# which should be the same plots as 
curto=readSBMLR(file.path(system.file(package="SBMLR"), "models/curto.r")) 
out1=simulate(curto,seq(-20,0,1)) 
curto$species$PRPP$ic=50 
out2=simulate(curto,0:70) 
outs=data.frame(rbind(out1,out2)) 

attach(outs) 
par(mfrow=c(2,1)) 
plot(time,IMP,type="l") 
plot(time,HX,type="l") 
par(mfrow=c(1,1)) 
detach(outs)

##---- The following example uses fderiv to generate Morrison's folate system response to 
morr=readSBMLR(file.path(system.file(package="SBMLR"), "models/morrison.r")) 
out1=simulate(morr,seq(-20,0,1)) 
morr$species$EMTX$ic=1 
out2=simulate(morr,0:30) 
outs=data.frame(rbind(out1,out2)) 

attach(outs)
pdf(file="output.pdf",paper="A4")
par(mfrow=c(3,4)) 
plot(time,FH2b,type="l",xlab="Hours") 
plot(time,FH2f,type="l",xlab="Hours") 
plot(time,DHFRf,type="l",xlab="Hours") 
plot(time,DHFRtot,type="l",xlab="Hours") 
plot(time,CHOFH4,type="l",xlab="Hours") 
plot(time,FH4,type="l",xlab="Hours") 
plot(time,CH2FH4,type="l",xlab="Hours") 
plot(time,CH3FH4,type="l",xlab="Hours") 
plot(time,AICARsyn,type="l",xlab="Hours") 
plot(time,MTR,type="l",xlab="Hours") 
plot(time,TYMS,type="l",xlab="Hours") 
#plot(time,EMTX,type="l",xlab="Hours") 
plot(time,DHFReductase,type="l",xlab="Hours")
dev.off()