Create and assess Create and assess protein networks protein networks through molecular through molecular characteristics of characteristics of individual proteins individual proteins Yanay Ofran et al. ISMB ’06 Yanay Ofran et al. ISMB ’06 Presenter: Danhua Guo Presenter: Danhua Guo 12/07/2006 12/07/2006
15
Embed
Create and assess protein networks through molecular characteristics of individual proteins
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Create and assess Create and assess protein networks protein networks through molecular through molecular characteristics of characteristics of individual proteinsindividual proteinsYanay Ofran et al. ISMB ’06Yanay Ofran et al. ISMB ’06
NetworkNetwork– Help identify process or functionsHelp identify process or functions– Major problemMajor problem
Generation problemGeneration problem– Experimental errors: should not be in the networkExperimental errors: should not be in the network– ““In vitroIn vitro”: should be include in the network”: should be include in the network
Data representation problemData representation problem– Essential connection between PPI and proteinEssential connection between PPI and protein
IntroductionIntroduction An ideal frameworkAn ideal framework
– Macro level: network topologyMacro level: network topology– Micro level: characteristics of each Micro level: characteristics of each
IntroductionIntroduction Protein interaction Network Protein interaction Network
Assessment Tool (PiNAT)Assessment Tool (PiNAT)
MethodsMethods Large-scale Assessment of PPIsLarge-scale Assessment of PPIs
– Based on localizationBased on localization– Based on GO annotation (if applicable)Based on GO annotation (if applicable)
Automatic generation of networksAutomatic generation of networks– Get submitted list of proteins from userGet submitted list of proteins from user– Search DIP and IntActSearch DIP and IntAct
Display of networks in the cellular Display of networks in the cellular contextcontext
Alzheimer’s disease related pathwayAlzheimer’s disease related pathway
– PHDhtm: predict transmembrane helices PHDhtm: predict transmembrane helices (7%)(7%) Threshold: average score among 20 reliable Threshold: average score among 20 reliable
predictions >8.5predictions >8.5– Experiment on 4800 interactions (2191 Experiment on 4800 interactions (2191
proteins)proteins) High-confidence prediction: 2312 (1482 proteins)High-confidence prediction: 2312 (1482 proteins) Total protein pairs: 1,097,421Total protein pairs: 1,097,421 Binomial approximation to the cumulative Binomial approximation to the cumulative
hypergeometric probability distribution to get a p-hypergeometric probability distribution to get a p-value for over and under representationvalue for over and under representation
MethodsMethods GO criteriaGO criteria
– The functionality annotation of a proteinThe functionality annotation of a protein– Distance between 2 GO terms measure the Distance between 2 GO terms measure the
similaritysimilarity
m,n: respective numbers of annotations in i and jm,n: respective numbers of annotations in i and j simGo: GO similarity defined by Lord et al.simGo: GO similarity defined by Lord et al. Ck, Cp: respective individual annotation in protein Ck, Cp: respective individual annotation in protein
i and ji and j Cjmax: Ck’s most similar term in jCjmax: Ck’s most similar term in j Cimax: Cp’s most similar term in iCimax: Cp’s most similar term in i
MethodsMethods Display of networks in the cellular Display of networks in the cellular
contextcontext– Based on LOCtree and PHDhtm Based on LOCtree and PHDhtm
predictionspredictions– Generate Graph Markup Language Generate Graph Markup Language