SlideShare a Scribd company logo
1 of 29
Download to read offline
Limitation of GWAS or Linkage analysis 
Introduction of WGP 
Lasso estimation 
Bayesian inference of Lasso 
Whole Genome Prediction Using Penalized 
Regression 
Bayesian Lasso 
Jinseob Kim, MD, MPH 
GSPH, SNU 
February 27, 2014 
Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
Limitation of GWAS or Linkage analysis 
Introduction of WGP 
Lasso estimation 
Bayesian inference of Lasso 
Contents 
1 Limitation of GWAS or Linkage analysis 
2 Introduction of WGP 
© 
3 Lasso estimation 
4 Bayesian inference of Lasso 
Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
Limitation of GWAS or Linkage analysis 
Introduction of WGP 
Lasso estimation 
Bayesian inference of Lasso 
Limitation 
1 GWAS & Linkage : No consistent result ! poor prediction. 
2 Complex traits : Overall eect (e.g:cardiovascular, cancer, 
etc..). 
Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
Limitation of GWAS or Linkage analysis 
Introduction of WGP 
Lasso estimation 
Bayesian inference of Lasso 
Example of GWAS 
GWASÐ ìì genetic informationD  ¨Ð ìhܤ” 
ƒ@ ˆ¥. 
1 1 trait VS 1 locus ! SNP /Ì| µÄÉ lä. 
2 Ș ´ SNP informationD à$t| XÀ multiple 
comparison p-value| t©XŒ ä. (ex: p-value cuto- 
5  108) 
3 Signi
cant SNPÌD Á ¨D l1 or combine 
information via Genetic Risk Score 
Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
Limitation of GWAS or Linkage analysis 
Introduction of WGP 
Lasso estimation 
Bayesian inference of Lasso 
Problem 
1 Multiple comparison ! Power... 
2 SNP X˜) trait@ „ ! LD information... 
3 What is Genetic Risk Score??? €U À.. 
Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
Limitation of GWAS or Linkage analysis 
Introduction of WGP 
Lasso estimation 
Bayesian inference of Lasso 
Why this problem? 
øå ä #à ŒÀ„Xt H L??? 
1 Multicolinearity issue!!! ! LD: similar allele information 
2 n  p issue: ‰, ¬Œôä À(SNP)/ Ît 
ŒÀÄ ”t H(. 
ŒÀÄX „°(variance)t 4 äÄä..... ”ˆ.. 
Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
Limitation of GWAS or Linkage analysis 
Introduction of WGP 
Lasso estimation 
Bayesian inference of Lasso 
Why?
”ÉX unbiaseness| ì0XÀ JX0 L8tä. 
Variance-bias trade-o!! 
(a) (b) 
Figure : Summary of variance-bias tradeo 
Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
Limitation of GWAS or Linkage analysis 
Introduction of WGP 
Lasso estimation 
Bayesian inference of Lasso 
Variance-bias tradeo 
Y = f (x) + ,   N(0; e ), ^f : estimate of f | L 
Err (x) = E[(Y  ^f (x))2] (1) 
Err (x) = (E[^f (x)  f (x)])2 + E[^f (x)  E[^f (x)]]2 + e (2) 
Err (x) = Bias2 + Variance + Irreducible error (3) 
Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
Limitation of GWAS or Linkage analysis 
Introduction of WGP 
Lasso estimation 
Bayesian inference of Lasso 
© 
Core context of WGP
unbiased estimator „D ì0ä!!! 
1 WGP can use all available markers to regress phenotype onto 
genomic information. 
Ridge regression 
Lasso (Least absolute shrinkage and selection operator) 
Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
Limitation of GWAS or Linkage analysis 
Introduction of WGP 
Lasso estimation 
Bayesian inference of Lasso 
© 
1 Lasso@ Bayesian inferenceX uì Ь@ Dt´Ì Lt 
ä. 
2 Lasso (¤À| t©X0  pt0 ¬| `  ˆä. 
3 „@  ÙT !  Ltü ø¼ Ý1. 
4 Data@ phenotype …% ! |8Ð ]` Ltü ø¼!! 
Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
Limitation of GWAS or Linkage analysis 
Introduction of WGP 
Lasso estimation 
Bayesian inference of Lasso 
Ridge VS Lasso 
Ridge regression 
minimize (~y  X
)T (~y  X
) s.t 
Xp 
j=1
2 
j  t 
$ minimize (~y  X
)T (~y  X
) +  
Xp 
j=1
2 
j 
(4) 
Lasso 
minimize (~y  X
)T (~y  X
) s.t 
Xp 
j=1 
j
j j  t 
$ minimize (~y  X
)T (~y  X
) +  
Xp 
j=1 
j
j j 
(5) 
Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
Limitation of GWAS or Linkage analysis 
Introduction of WGP 
Lasso estimation 
Bayesian inference of Lasso 
Ridge VS Lasso(2) 
1 P )• ¨P Î@ betaäD 0 ô¸ä. äõ 1 
t°, LD information . 
2 Square(
2) VS Abs(j
j) 
3 0.04 VS 0.2 : ñt ôä ‘@

More Related Content

More from Jinseob Kim

Unsupervised Deep Learning Applied to Breast Density Segmentation and Mammogr...
Unsupervised Deep Learning Applied to Breast Density Segmentation and Mammogr...Unsupervised Deep Learning Applied to Breast Density Segmentation and Mammogr...
Unsupervised Deep Learning Applied to Breast Density Segmentation and Mammogr...Jinseob Kim
 
Fst, selection index
Fst, selection indexFst, selection index
Fst, selection indexJinseob Kim
 
Why Does Deep and Cheap Learning Work So Well
Why Does Deep and Cheap Learning Work So WellWhy Does Deep and Cheap Learning Work So Well
Why Does Deep and Cheap Learning Work So WellJinseob Kim
 
괴델(Godel)의 불완전성 정리 증명의 이해.
괴델(Godel)의 불완전성 정리 증명의 이해.괴델(Godel)의 불완전성 정리 증명의 이해.
괴델(Godel)의 불완전성 정리 증명의 이해.Jinseob Kim
 
New Epidemiologic Measures in Multilevel Study: Median Risk Ratio, Median Haz...
New Epidemiologic Measures in Multilevel Study: Median Risk Ratio, Median Haz...New Epidemiologic Measures in Multilevel Study: Median Risk Ratio, Median Haz...
New Epidemiologic Measures in Multilevel Study: Median Risk Ratio, Median Haz...Jinseob Kim
 
가설검정의 심리학
가설검정의 심리학 가설검정의 심리학
가설검정의 심리학 Jinseob Kim
 
Win Above Replacement in Sabermetrics
Win Above Replacement in SabermetricsWin Above Replacement in Sabermetrics
Win Above Replacement in SabermetricsJinseob Kim
 
Regression Basic : MLE
Regression  Basic : MLERegression  Basic : MLE
Regression Basic : MLEJinseob Kim
 
iHS calculation in R
iHS calculation in RiHS calculation in R
iHS calculation in RJinseob Kim
 
Selection index population_genetics
Selection index population_geneticsSelection index population_genetics
Selection index population_geneticsJinseob Kim
 
질병부담계산: Dismod mr gbd2010
질병부담계산: Dismod mr gbd2010질병부담계산: Dismod mr gbd2010
질병부담계산: Dismod mr gbd2010Jinseob Kim
 
Case-crossover study
Case-crossover studyCase-crossover study
Case-crossover studyJinseob Kim
 
Generalized Additive Model
Generalized Additive Model Generalized Additive Model
Generalized Additive Model Jinseob Kim
 
Deep Learning by JSKIM (Korean)
Deep Learning by JSKIM (Korean)Deep Learning by JSKIM (Korean)
Deep Learning by JSKIM (Korean)Jinseob Kim
 
Machine Learning Introduction
Machine Learning IntroductionMachine Learning Introduction
Machine Learning IntroductionJinseob Kim
 
Deep learning by JSKIM
Deep learning by JSKIMDeep learning by JSKIM
Deep learning by JSKIMJinseob Kim
 

More from Jinseob Kim (20)

Unsupervised Deep Learning Applied to Breast Density Segmentation and Mammogr...
Unsupervised Deep Learning Applied to Breast Density Segmentation and Mammogr...Unsupervised Deep Learning Applied to Breast Density Segmentation and Mammogr...
Unsupervised Deep Learning Applied to Breast Density Segmentation and Mammogr...
 
Fst, selection index
Fst, selection indexFst, selection index
Fst, selection index
 
Why Does Deep and Cheap Learning Work So Well
Why Does Deep and Cheap Learning Work So WellWhy Does Deep and Cheap Learning Work So Well
Why Does Deep and Cheap Learning Work So Well
 
괴델(Godel)의 불완전성 정리 증명의 이해.
괴델(Godel)의 불완전성 정리 증명의 이해.괴델(Godel)의 불완전성 정리 증명의 이해.
괴델(Godel)의 불완전성 정리 증명의 이해.
 
New Epidemiologic Measures in Multilevel Study: Median Risk Ratio, Median Haz...
New Epidemiologic Measures in Multilevel Study: Median Risk Ratio, Median Haz...New Epidemiologic Measures in Multilevel Study: Median Risk Ratio, Median Haz...
New Epidemiologic Measures in Multilevel Study: Median Risk Ratio, Median Haz...
 
가설검정의 심리학
가설검정의 심리학 가설검정의 심리학
가설검정의 심리학
 
Win Above Replacement in Sabermetrics
Win Above Replacement in SabermetricsWin Above Replacement in Sabermetrics
Win Above Replacement in Sabermetrics
 
Regression Basic : MLE
Regression  Basic : MLERegression  Basic : MLE
Regression Basic : MLE
 
iHS calculation in R
iHS calculation in RiHS calculation in R
iHS calculation in R
 
Fst in R
Fst in R Fst in R
Fst in R
 
Selection index population_genetics
Selection index population_geneticsSelection index population_genetics
Selection index population_genetics
 
질병부담계산: Dismod mr gbd2010
질병부담계산: Dismod mr gbd2010질병부담계산: Dismod mr gbd2010
질병부담계산: Dismod mr gbd2010
 
DALY & QALY
DALY & QALYDALY & QALY
DALY & QALY
 
Case-crossover study
Case-crossover studyCase-crossover study
Case-crossover study
 
Generalized Additive Model
Generalized Additive Model Generalized Additive Model
Generalized Additive Model
 
Deep Learning by JSKIM (Korean)
Deep Learning by JSKIM (Korean)Deep Learning by JSKIM (Korean)
Deep Learning by JSKIM (Korean)
 
Machine Learning Introduction
Machine Learning IntroductionMachine Learning Introduction
Machine Learning Introduction
 
Tree advanced
Tree advancedTree advanced
Tree advanced
 
Deep learning by JSKIM
Deep learning by JSKIMDeep learning by JSKIM
Deep learning by JSKIM
 
Main result
Main result Main result
Main result
 

Recently uploaded

Conf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming PipelinesConf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming PipelinesTimothy Spann
 
What To Do For World Nature Conservation Day by Slidesgo.pptx
What To Do For World Nature Conservation Day by Slidesgo.pptxWhat To Do For World Nature Conservation Day by Slidesgo.pptx
What To Do For World Nature Conservation Day by Slidesgo.pptxSimranPal17
 
INTRODUCTION TO Natural language processing
INTRODUCTION TO Natural language processingINTRODUCTION TO Natural language processing
INTRODUCTION TO Natural language processingsocarem879
 
Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...
Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...
Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...Thomas Poetter
 
Principles and Practices of Data Visualization
Principles and Practices of Data VisualizationPrinciples and Practices of Data Visualization
Principles and Practices of Data VisualizationKianJazayeri1
 
Defining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data StoryDefining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data StoryJeremy Anderson
 
Networking Case Study prepared by teacher.pptx
Networking Case Study prepared by teacher.pptxNetworking Case Study prepared by teacher.pptx
Networking Case Study prepared by teacher.pptxHimangsuNath
 
modul pembelajaran robotic Workshop _ by Slidesgo.pptx
modul pembelajaran robotic Workshop _ by Slidesgo.pptxmodul pembelajaran robotic Workshop _ by Slidesgo.pptx
modul pembelajaran robotic Workshop _ by Slidesgo.pptxaleedritatuxx
 
Student Profile Sample report on improving academic performance by uniting gr...
Student Profile Sample report on improving academic performance by uniting gr...Student Profile Sample report on improving academic performance by uniting gr...
Student Profile Sample report on improving academic performance by uniting gr...Seán Kennedy
 
Bank Loan Approval Analysis: A Comprehensive Data Analysis Project
Bank Loan Approval Analysis: A Comprehensive Data Analysis ProjectBank Loan Approval Analysis: A Comprehensive Data Analysis Project
Bank Loan Approval Analysis: A Comprehensive Data Analysis ProjectBoston Institute of Analytics
 
Unveiling the Role of Social Media Suspect Investigators in Preventing Online...
Unveiling the Role of Social Media Suspect Investigators in Preventing Online...Unveiling the Role of Social Media Suspect Investigators in Preventing Online...
Unveiling the Role of Social Media Suspect Investigators in Preventing Online...Milind Agarwal
 
Advanced Machine Learning for Business Professionals
Advanced Machine Learning for Business ProfessionalsAdvanced Machine Learning for Business Professionals
Advanced Machine Learning for Business ProfessionalsVICTOR MAESTRE RAMIREZ
 
Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...
Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...
Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...Boston Institute of Analytics
 
Predictive Analysis for Loan Default Presentation : Data Analysis Project PPT
Predictive Analysis for Loan Default  Presentation : Data Analysis Project PPTPredictive Analysis for Loan Default  Presentation : Data Analysis Project PPT
Predictive Analysis for Loan Default Presentation : Data Analysis Project PPTBoston Institute of Analytics
 
Cyber awareness ppt on the recorded data
Cyber awareness ppt on the recorded dataCyber awareness ppt on the recorded data
Cyber awareness ppt on the recorded dataTecnoIncentive
 
FAIR, FAIRsharing, FAIR Cookbook and ELIXIR - Sansone SA - Boston 2024
FAIR, FAIRsharing, FAIR Cookbook and ELIXIR - Sansone SA - Boston 2024FAIR, FAIRsharing, FAIR Cookbook and ELIXIR - Sansone SA - Boston 2024
FAIR, FAIRsharing, FAIR Cookbook and ELIXIR - Sansone SA - Boston 2024Susanna-Assunta Sansone
 
Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...
Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...
Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...Boston Institute of Analytics
 
convolutional neural network and its applications.pdf
convolutional neural network and its applications.pdfconvolutional neural network and its applications.pdf
convolutional neural network and its applications.pdfSubhamKumar3239
 
Real-Time AI Streaming - AI Max Princeton
Real-Time AI  Streaming - AI Max PrincetonReal-Time AI  Streaming - AI Max Princeton
Real-Time AI Streaming - AI Max PrincetonTimothy Spann
 

Recently uploaded (20)

Conf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming PipelinesConf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
 
What To Do For World Nature Conservation Day by Slidesgo.pptx
What To Do For World Nature Conservation Day by Slidesgo.pptxWhat To Do For World Nature Conservation Day by Slidesgo.pptx
What To Do For World Nature Conservation Day by Slidesgo.pptx
 
INTRODUCTION TO Natural language processing
INTRODUCTION TO Natural language processingINTRODUCTION TO Natural language processing
INTRODUCTION TO Natural language processing
 
Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...
Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...
Minimizing AI Hallucinations/Confabulations and the Path towards AGI with Exa...
 
Principles and Practices of Data Visualization
Principles and Practices of Data VisualizationPrinciples and Practices of Data Visualization
Principles and Practices of Data Visualization
 
Defining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data StoryDefining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data Story
 
Networking Case Study prepared by teacher.pptx
Networking Case Study prepared by teacher.pptxNetworking Case Study prepared by teacher.pptx
Networking Case Study prepared by teacher.pptx
 
modul pembelajaran robotic Workshop _ by Slidesgo.pptx
modul pembelajaran robotic Workshop _ by Slidesgo.pptxmodul pembelajaran robotic Workshop _ by Slidesgo.pptx
modul pembelajaran robotic Workshop _ by Slidesgo.pptx
 
Student Profile Sample report on improving academic performance by uniting gr...
Student Profile Sample report on improving academic performance by uniting gr...Student Profile Sample report on improving academic performance by uniting gr...
Student Profile Sample report on improving academic performance by uniting gr...
 
Bank Loan Approval Analysis: A Comprehensive Data Analysis Project
Bank Loan Approval Analysis: A Comprehensive Data Analysis ProjectBank Loan Approval Analysis: A Comprehensive Data Analysis Project
Bank Loan Approval Analysis: A Comprehensive Data Analysis Project
 
Unveiling the Role of Social Media Suspect Investigators in Preventing Online...
Unveiling the Role of Social Media Suspect Investigators in Preventing Online...Unveiling the Role of Social Media Suspect Investigators in Preventing Online...
Unveiling the Role of Social Media Suspect Investigators in Preventing Online...
 
Advanced Machine Learning for Business Professionals
Advanced Machine Learning for Business ProfessionalsAdvanced Machine Learning for Business Professionals
Advanced Machine Learning for Business Professionals
 
Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...
Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...
Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...
 
Predictive Analysis for Loan Default Presentation : Data Analysis Project PPT
Predictive Analysis for Loan Default  Presentation : Data Analysis Project PPTPredictive Analysis for Loan Default  Presentation : Data Analysis Project PPT
Predictive Analysis for Loan Default Presentation : Data Analysis Project PPT
 
Cyber awareness ppt on the recorded data
Cyber awareness ppt on the recorded dataCyber awareness ppt on the recorded data
Cyber awareness ppt on the recorded data
 
FAIR, FAIRsharing, FAIR Cookbook and ELIXIR - Sansone SA - Boston 2024
FAIR, FAIRsharing, FAIR Cookbook and ELIXIR - Sansone SA - Boston 2024FAIR, FAIRsharing, FAIR Cookbook and ELIXIR - Sansone SA - Boston 2024
FAIR, FAIRsharing, FAIR Cookbook and ELIXIR - Sansone SA - Boston 2024
 
Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...
Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...
Data Analysis Project : Targeting the Right Customers, Presentation on Bank M...
 
convolutional neural network and its applications.pdf
convolutional neural network and its applications.pdfconvolutional neural network and its applications.pdf
convolutional neural network and its applications.pdf
 
Real-Time AI Streaming - AI Max Princeton
Real-Time AI  Streaming - AI Max PrincetonReal-Time AI  Streaming - AI Max Princeton
Real-Time AI Streaming - AI Max Princeton
 
Data Analysis Project: Stroke Prediction
Data Analysis Project: Stroke PredictionData Analysis Project: Stroke Prediction
Data Analysis Project: Stroke Prediction
 

Whole Genome Regression using Bayesian Lasso

  • 1. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Whole Genome Prediction Using Penalized Regression Bayesian Lasso Jinseob Kim, MD, MPH GSPH, SNU February 27, 2014 Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 2. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Contents 1 Limitation of GWAS or Linkage analysis 2 Introduction of WGP © 3 Lasso estimation 4 Bayesian inference of Lasso Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 3. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Limitation 1 GWAS & Linkage : No consistent result ! poor prediction. 2 Complex traits : Overall eect (e.g:cardiovascular, cancer, etc..). Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 4. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Example of GWAS GWASÐ ìì genetic informationD ¨Ð ìhܤ” ƒ@ ˆ¥. 1 1 trait VS 1 locus ! SNP /Ì| µÄÉ lä. 2 Ș ´ SNP informationD à$t| XÀ multiple comparison p-value| t©XŒ ä. (ex: p-value cuto- 5 108) 3 Signi
  • 5. cant SNPÌD Á ¨D l1 or combine information via Genetic Risk Score Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 6. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Problem 1 Multiple comparison ! Power... 2 SNP X˜) trait@ „ ! LD information... 3 What is Genetic Risk Score??? €U À.. Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 7. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Why this problem? øå ä #à ŒÀ„Xt H L??? 1 Multicolinearity issue!!! ! LD: similar allele information 2 n p issue: ‰, ¬Œôä À(SNP)/ Ît ŒÀÄ ”t H(. ŒÀÄX „°(variance)t 4 äÄä..... ”ˆ.. Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 8. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Why?
  • 9. ”ÉX unbiaseness| ì0XÀ JX0 L8tä. Variance-bias trade-o!! (a) (b) Figure : Summary of variance-bias tradeo Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 10. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Variance-bias tradeo Y = f (x) + , N(0; e ), ^f : estimate of f | L Err (x) = E[(Y ^f (x))2] (1) Err (x) = (E[^f (x) f (x)])2 + E[^f (x) E[^f (x)]]2 + e (2) Err (x) = Bias2 + Variance + Irreducible error (3) Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 11. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso © Core context of WGP
  • 12. unbiased estimator „D ì0ä!!! 1 WGP can use all available markers to regress phenotype onto genomic information. Ridge regression Lasso (Least absolute shrinkage and selection operator) Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 13. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso © 1 Lasso@ Bayesian inferenceX uì Ь@ Dt´Ì Lt ä. 2 Lasso (¤À| t©X0 pt0 ¬| ` ˆä. 3 „@ ÙT ! Ltü ø¼ Ý1. 4 Data@ phenotype …% ! |8Ð ]` Ltü ø¼!! Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 14. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Ridge VS Lasso Ridge regression minimize (~y X
  • 15. )T (~y X
  • 16. ) s.t Xp j=1
  • 17. 2 j t $ minimize (~y X
  • 18. )T (~y X
  • 19. ) + Xp j=1
  • 20. 2 j (4) Lasso minimize (~y X
  • 21. )T (~y X
  • 22. ) s.t Xp j=1 j
  • 23. j j t $ minimize (~y X
  • 24. )T (~y X
  • 25. ) + Xp j=1 j
  • 26. j j (5) Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 27. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Ridge VS Lasso(2) 1 P )• ¨P Î@ betaäD 0 ô¸ä. äõ 1 t°, LD information . 2 Square(
  • 29. j) 3 0.04 VS 0.2 : ñt ôä ‘@
  • 30. D 0 T ˜ ô¸ä. 4 t T pt, ‰ T Î@
  • 31. äD 0 ô¸ä. 5 Lasso ridgeôä T Î@
  • 32. äD 0 ô¸ä. Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 33. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Ridge VS Lasso(3) Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 34. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Choosing K-fold cross validation pt0Ð k € n käD Àà Modeling Ä tƒD kX sampleÐ ©Xì error l Ä øƒäD ä Éà ƒD CV error| ä. CV erroräX ÉàD ŒT X” lä. (CV error)() = E((CV error)() k ) (6) Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 35. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso 10 fold CV Figure : 10 fold CV Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 36. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Bayesian inference Introduction ! ThinkBayes X] gogo!! Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 37. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Lasso ! Bayesian Lasso
  • 38. i j2 2 ej
  • 39. i j= : Laplace prior 1 The Laplacian prior assigns more weight to regions near zero than the normal prior. 2 Interpretated as mixture of the hierarchical priors (Normal + exponential) a2 eajzj = R 1 0 p1 2s ez2=2s a2 2 eas2=2ds, a 0 Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 40. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Laplace prior Figure : Normal VS Laplace prior Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 41. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Example: Continuous case Whole model i = + XJ j=1 xij j + XL l=1 zlj
  • 42. j (7) Likelihood p(yi ji ; 2) = (22)1 2 expf (yi i )2 22 g (8) Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 43. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Likelihood p(yj; ;
  • 44. ; 2) = Y N(yi ji + XJ j=1 xij j + XL l=1 zij
  • 45. j ; 2) (9) y = fyig; = f jg;
  • 46. = f
  • 47. lg Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 48. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Prior construction: hierarchial model 1 Intercept() sex, smoking, BMI( ) : vague prior(non-informative) 2 Residual variance - standard assumption of bayesian regression : scaled-inverse Chi-square density 2(2jdf ; S) 3 marker eect - bayesian Lasso p(
  • 49. ; Q 2; 2jH; 2) = p(
  • 50. j 22)p( 2j2)p(2jr ; s) = f L l=1 N(
  • 51. l j0; 2 l 2)Exp( 2 l j2)gG(2jr ; s) Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 52. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Prior p(; ; 2;
  • 53. ; 2; 2jH) / 2(2jdf ; S)f YL l=1 N(
  • 54. l j0; 2 l 2)Exp( 2 l j2)gG(2jr ; s) (10) H = fdf = 5; S = 170; = 1 104; s = 2g : For priors with small in uences on predictions Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 55. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Posterior p(; ; 2;
  • 56. ; 2; 2jy) / Y N(yi ji + XJ j=1 xij j + XL l=1 zij
  • 57. j ; 2) YL 2(2jdf ; S)f l=1 N(
  • 58. l j0; 2 l 2)Exp( 2 l j2)gG(2jr ; s) (11) Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 59. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Implementation 1 BLR(Bayesian Linear Regression) package in R 2 bayesm, splines and SuppDists for sampler ! BGLR(Bayesian Generalized Linear Regression) package in R Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 60. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso Goodness of
  • 61. t, DIC Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 62. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso äµ BGLR package äµ : continuous trait (TG) binomial traint (hyperTG) Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 63. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso üX¬m 1 ø¬ À1È(conti VS categorial) À. 2 Lasso ø À(genotype)@ øå À(age)| l„ 3 t` Lasso Ð ä´ xä@ ¨P T´| ä.  Àt õÉXŒ !´| X0 L8tä. Ș allele count” 4pt 0,1,2tÀ ÁÆL. 4 Missingt Æ´| ä. GWAS” Missing |à LD Ä°tüÀÌ BGLR@ øÀ Jä. Œä prediction modeltÀ TT± xÐ missing Æ´| h: Imputation or mean allele count. Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 64. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso üX¬m2 Validation ` ƒt|t 1 P SetX õµ SNPÌ !¨ l1Xì| ä. 2 P SetX allele count reference Ù|Xì| ä. 3 P SetÐ ¨P tù traitt ˆ´| ä. Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression
  • 65. Limitation of GWAS or Linkage analysis Introduction of WGP Lasso estimation Bayesian inference of Lasso ] HP: 010-9192-5385 E-mail: secondmath85@gmail.com Jinseob Kim, MD, MPH Whole Genome Prediction Using Penalized Regression