Flexible linear models John Blischak Instructor DataCamp - PowerPoint PPT Presentation

DataCamp Differential Expression Analysis with limma in R DIFFERENTIAL EXPRESSION ANALYSIS WITH LIMMA IN R Flexible linear models John Blischak Instructor

DataCamp Differential Expression Analysis with limma in R Models for complicated study designs Y = β + β X + ϵ Y = β + β X + β X + ϵ 0 1 1 0 1 1 2 2 β - Mean in ER-neg β - Mean in group 1 0 0 β - Mean difference in ER-pos β - Mean difference in group 2 1 1 Test: β = 0 β - Mean difference in group 3 1 2 Tests: β = 0 , β = 0 , ??? 1 2

DataCamp Differential Expression Analysis with limma in R Group-means parametrization Y = β X + β X + ϵ Y = β X + β X + β X + ϵ 1 1 2 2 1 1 2 2 3 3 β - Mean in ER-neg β - Mean in group 1 1 1 β - Mean in ER-pos β - Mean in group 2 2 2 Test: β − β = 0 β - Mean in group 3 2 1 3 Tests: β − β = 0 2 1 β − β = 0 3 1 β − β = 0 3 2

DataCamp Differential Expression Analysis with limma in R Design matrix for group-means design <- model.matrix(~0 + er, data = pData(eset)) head(design) ernegative erpositive VDX_3 1 0 VDX_5 0 1 VDX_6 1 0 VDX_7 1 0 VDX_8 1 0 VDX_9 0 1 colSums(design) ernegative erpositive 135 209

DataCamp Differential Expression Analysis with limma in R Contrasts matrix library(limma) cm <- makeContrasts(status = erpositive - ernegative, levels = design) cm Contrasts Levels status ernegative -1 erpositive 1

DataCamp Differential Expression Analysis with limma in R Testing the group-means parametrization fit <- lmFit(eset, design) head(fit$coefficients, 3) ernegative erpositive 1007_s_at 11.725148 11.823936 1053_at 8.126934 7.580204 117_at 7.972049 7.798623 fit2 <- contrasts.fit(fit, contrasts = cm) head(fit2$coefficients, 3) Contrasts status 1007_s_at 0.09878782 1053_at -0.54673000 117_at -0.17342654

DataCamp Differential Expression Analysis with limma in R The parametrization does not change the results # Calculate the t-statistics fit2 <- eBayes(fit2) # Count the number of differentially expressed genes results <- decideTests(fit2) summary(results) status -1 6276 0 11003 1 5004

DataCamp Differential Expression Analysis with limma in R DIFFERENTIAL EXPRESSION ANALYSIS WITH LIMMA IN R Let's practice!

DataCamp Differential Expression Analysis with limma in R DIFFERENTIAL EXPRESSION ANALYSIS WITH LIMMA IN R Studies with more than two groups John Blischak Instructor

DataCamp Differential Expression Analysis with limma in R A study with 3 groups 3 different types of leukemias: ALL, AML, CML Bioconductor package: leukemiasEset Kohlmann et al. 2008, Haferlach et al. 2010 dim(eset) Features Samples 20172 36 table(pData(eset)[, "type"]) ALL AML CML 12 12 12

DataCamp Differential Expression Analysis with limma in R Group-means model for 3 groups Y = β X + β X + β X + ϵ 1 1 2 2 3 3 β - Mean expression level in group ALL 1 β - Mean expression level in group AML 2 β - Mean expression level in group CML 3 Tests: AML v. ALL: β − β = 0 2 1 CML v. ALL: β − β = 0 3 1 CML v. AML: β − β = 0 3 2

DataCamp Differential Expression Analysis with limma in R Group-means design matrix for 3 groups design <- model.matrix(~0 + type, data = pData(eset)) head(design, 3) typeALL typeAML typeCML sample_01 1 0 0 sample_02 1 0 0 sample_03 1 0 0 colSums(design) typeALL typeAML typeCML 12 12 12

DataCamp Differential Expression Analysis with limma in R Contrasts matrix for 3 groups AML v. ALL: β − β = 0 2 1 CML v. ALL: β − β = 0 3 1 CML v. AML: β − β = 0 3 2 library(limma) cm <- makeContrasts(AMLvALL = typeAML - typeALL, CMLvALL = typeCML - typeALL, CMLvAML = typeCML - typeAML, levels = design) cm Contrasts Levels AMLvALL CMLvALL CMLvAML typeALL -1 -1 0 typeAML 1 0 -1 typeCML 0 1 1

DataCamp Differential Expression Analysis with limma in R Testing 3 groups library(limma) # Fit coefficients fit <- lmFit(eset, design) # Fit contrasts fit2 <- contrasts.fit(fit, contrasts = cm) # Calculate t-statistics fit2 <- eBayes(fit2) # Summarize results results <- decideTests(fit2) summary(results) AMLvALL CMLvALL CMLvAML -1 898 3401 1890 0 18323 13194 16408 1 951 3577 1874

DataCamp Differential Expression Analysis with limma in R The effect of hypoxia on stem cell function 3 different levels of oxygen: 1%, 5%, 21% Bioconductor package: stemHypoxia Prado-Lopez et al. 2010 dim(eset) Features Samples 15325 6 table(pData(eset)[, "oxygen"]) ox01 ox05 ox21 2 2 2

DataCamp Differential Expression Analysis with limma in R DIFFERENTIAL EXPRESSION ANALYSIS WITH LIMMA IN R Factorial experimental design John Blischak Instructor

DataCamp Differential Expression Analysis with limma in R Factorial designs 2x2 design to study effect of low temperature in plants: 2 types of Arabidopsis thaliana : col, vte2 2 temperatures: normal, low Maeda et al. 2010 dim(eset) Features Samples 11871 12 table(pData(eset)[, c("type", "temp")]) temp type low normal col 3 3 vte2 3 3

DataCamp Differential Expression Analysis with limma in R Group-means model for 2x2 factorial Y = β X + β X + β X + β X + ϵ 1 1 2 2 3 3 4 4 β - Mean expression level in col plants at low temperature 1 β - Mean expression level in col plants at normal temperature 2 β - Mean expression level in vte2 plants at low temperature 3 β - Mean expression level in vte2 plants at normal temperature 4

DataCamp Differential Expression Analysis with limma in R Group-means design matrix for 2x2 factorial group <- with(pData(eset), paste(type, temp, sep = ".")) group <- factor(group) design <- model.matrix(~0 + group) colnames(design) <- levels(group) head(design, 3) col.low col.normal vte2.low vte2.normal 1 0 1 0 0 2 0 1 0 0 3 0 1 0 0 colSums(design) col.low col.normal vte2.low vte2.normal 3 3 3 3

DataCamp Differential Expression Analysis with limma in R Contrasts for a 2x2 factorial β β β β 1 2 3 4 col col vte2 vte2 type low normal low normal temp Differences of type in normal temp : β − β = 0 4 2 Differences of type in low temp : β − β = 0 3 1 Differences of temp in vte2 type : β − β = 0 3 4 Effect of temp in col type : β − β = 0 1 2 Differences of temp between col and vte2 type : ( β − β ) − ( β − β ) = 0 3 4 1 2

DataCamp Differential Expression Analysis with limma in R Contrasts matrix for 2x2 factorial library(limma) cm <- makeContrasts(type_normal = vte2.normal - col.normal, type_low = vte2.low - col.low, temp_vte2 = vte2.low - vte2.normal, temp_col = col.low - col.normal, interaction = (vte2.low - vte2.normal) - (col.low - col.normal), levels = design) cm Contrasts Levels type_normal type_low temp_vte2 temp_col interaction col.low 0 -1 0 1 -1 col.normal -1 0 0 -1 1 vte2.low 0 1 1 0 1 vte2.normal 1 0 -1 0 -1

DataCamp Differential Expression Analysis with limma in R Testing 2x2 factorial library(limma) # Fit coefficients fit <- lmFit(eset, design) # Fit contrasts fit2 <- contrasts.fit(fit, contrasts = cm) # Calculate t-statistics fit2 <- eBayes(fit2) # Summarize results results <- decideTests(fit2) summary(results) type_normal type_low temp_vte2 temp_col interaction -1 0 466 1635 1885 128 0 11871 10915 7635 6989 11640 1 0 490 2601 2997 103

DataCamp Differential Expression Analysis with limma in R The effect of drought on Populus trees 2x2 design to study effect of drought in trees: 2 types of Populus : DN34, NM6 2 water conditions: normal, drought Wilkins et al. 2009 dim(eset) Features Samples 16172 12 table(pData(eset)[, c("type", "water")]) water type drought normal dn34 3 3 nm6 3 3

Flexible linear models John Blischak Instructor DataCamp - PowerPoint PPT Presentation

DataCamp Differential Expression Analysis with limma in R DIFFERENTIAL EXPRESSION ANALYSIS WITH LIMMA IN R Flexible linear models John Blischak Instructor DataCamp Differential Expression Analysis with limma in R Models for complicated study

Neural Networks Still seeking flexible, non-linear models for classfication and CS 335: Neural

Linear mixed models with improper priors and flexible distributional assumptions for longitudinal

Introduction to Data Science: Logistic 0 1 1 according to a data fit criterion. account

Functional Linear Models 1 66 / 181 Functional Linear Models Statistical Models So far we have

Notes on the Non-linear Regression The model Non-linear regression models, like ordinary linear

Workshop 3 Building from Linear Models to Generalised Linear Models Part 2: GLMs 2 2 What are

Workshop 2 Building from Linear Models to Generalised Linear Models Part 1: understanding LMs 2

Limitations of linear models Richard Erickson Instructor DataCamp Generalized Linear Models in

Outline Statistical inference for linear mixed models general form of linear mixed models

Introduction to General and Generalized Linear Models General Linear Models - part I Henrik

Linear Models are Most Favorable among Generalized Linear Models Kuan-Yun Lee and Thomas A.

Why Threshold Models: Need to Go Beyond . . . The Above Idea Works . . . A Theoretical

Plan Dynamic Linear Models Bayesian Analysis of Dynamic Linear Models in R The R package dlm

ECON 950 Winter 2020 Prof. James MacKinnon 9. Going Beyond Linear Models Linear regression,

Flexible parametric joint modelling of longitudinal and survival data Workshop on Flexible Models

Extended multivariate generalised linear and non-linear mixed effects models Stata UK Meeting

Clustering by contrast Cyril CHHUN Tlcom Paris Advisor: Jean-Louis DESSALLES June 20, 2019

Introduction to statistics: Linear models Shravan Vasishth Universit at Potsdam

Contour integration methods for self adjoint operators OTKR December 1922 2019 TU Vienna

Astrophysical and Dark Matter Origin of the IceCube High-energy Neutrino Events B HUPAL D EV

Practical Interpretation With a quantitative factor, like power in the etch-rate example, typically

A cross-linguistic comparison of the L2 acquisition of VOT contrasts Katherine Zhang Carnegie

A learning strategy for contrast-agnostic segmentation of brain MRI scans Benjamin Billot Billot

Explaining at-issueness contrasts between questions and assertions Matthijs Westera Institute