Differential expression analysis Mary Piper Bioinformatics - - PowerPoint PPT Presentation

▶

differential expression analysis

Differential expression analysis Mary Piper Bioinformatics - - PowerPoint PPT Presentation

Nov 13, 2023 41 likes •507 views

DataCamp RNA-Seq Differential Expression Analysis RNA - SEQ DIFFERENTIAL EXPRESSION ANALYSIS Differential expression analysis Mary Piper Bioinformatics Consultant and Trainer DataCamp RNA-Seq Differential Expression Analysis DataCamp

slide-1

SLIDE 1

DataCamp RNA-Seq Differential Expression Analysis

Differential expression analysis

RNA-SEQ DIFFERENTIAL EXPRESSION ANALYSIS

Mary Piper

Bioinformatics Consultant and Trainer

slide-2

SLIDE 2

DataCamp RNA-Seq Differential Expression Analysis

slide-3

SLIDE 3

DataCamp RNA-Seq Differential Expression Analysis

Differential expression analysis: DESeq2 vignette

vignette(DESeq2)

slide-4

SLIDE 4

DataCamp RNA-Seq Differential Expression Analysis

slide-5

SLIDE 5

DataCamp RNA-Seq Differential Expression Analysis

slide-6

SLIDE 6

DataCamp RNA-Seq Differential Expression Analysis

Bringing in data for DESeq2

# Read in raw counts wt_rawcounts <- read.csv("fibrosis_wt_rawcounts.csv") View(wt_rawcounts)

slide-7

SLIDE 7

DataCamp RNA-Seq Differential Expression Analysis

Bringing in data for DESeq2: metadata

# Read in metadata wt_metadata <- read.csv("fibrosis_wt_metadata_unordered.csv") View(wt_metadata)

slide-8

SLIDE 8

DataCamp RNA-Seq Differential Expression Analysis

Let's practice!

RNA-SEQ DIFFERENTIAL EXPRESSION ANALYSIS

slide-9

SLIDE 9

DataCamp RNA-Seq Differential Expression Analysis

Organizing the data for DESeq2

RNA-SEQ DIFFERENTIAL EXPRESSION ANALYSIS

Mary Piper

Bioinformatics Consultant and Trainer

slide-10

SLIDE 10

DataCamp RNA-Seq Differential Expression Analysis

Bringing in data for DESeq2: sample order

Metadata Raw counts

slide-11

SLIDE 11

DataCamp RNA-Seq Differential Expression Analysis

Bringing in data for DESeq2: sample order

rownames(wt_metadata) [1] "wt_normal3" "smoc2_fibrosis2" "wt_fibrosis3" [4] "smoc2_fibrosis3" "smoc2_normal3" "wt_normal1" [7] "smoc2_normal4" "wt_fibrosis2" "wt_normal2" [10] "smoc2_normal1" "smoc2_fibrosis1" "smoc2_fibrosis4" [13] "wt_fibrosis4" "wt_fibrosis1" colnames(wt_rawcounts) [1] "wt_normal1" "wt_normal2" "wt_normal3" [4] "wt_fibrosis1" "wt_fibrosis2" "wt_fibrosis3" [7] "wt_fibrosis4" "smoc2_normal1" "smoc2_normal3" [10] "smoc2_normal4" "smoc2_fibrosis1" "smoc2_fibrosis2" [13] "smoc2_fibrosis3" "smoc2_fibrosis4"

slide-12

SLIDE 12

DataCamp RNA-Seq Differential Expression Analysis

Bringing in data for DESeq2: sample order

all(rownames(wt_metadata) == colnames(wt_rawcounts)) [1] FALSE

slide-13

SLIDE 13

DataCamp RNA-Seq Differential Expression Analysis

Matching order between vectors

Using the match() function: vector1: vector of values with the desired order vector2: vector of values to reorder

utput: the indices for how to rearrange vector2 to be in the same order as vector1

match(vector1, vector2) match(colnames(wt_rawcounts), rownames(wt_metadata) [1] 6 9 1 14 8 3 [7] 13 10 5 7 11 2 [13] 4 12

slide-14

SLIDE 14

DataCamp RNA-Seq Differential Expression Analysis

Reordering with the match() function

Reordering using match() output:

idx <- match(colnames(wt_rawcounts), rownames(wt_metadata)) reordered_wt_metadata <- wt_metadata[idx, ] View(reordered_wt_metadata)

slide-15

SLIDE 15

DataCamp RNA-Seq Differential Expression Analysis

Checking the order

all(rownames(reordered_wt_metadata) == colnames(wt_rawcounts)) [1] TRUE

slide-16

SLIDE 16

DataCamp RNA-Seq Differential Expression Analysis

Creating the DESeq2 object

# Create DESeq object dds_wt <- DESeqDataSetFromMatrix(countData = wt_rawcounts, colData = reordered_wt_metadata, design = ~ condition)

slide-17

SLIDE 17

DataCamp RNA-Seq Differential Expression Analysis

Let's practice!

RNA-SEQ DIFFERENTIAL EXPRESSION ANALYSIS

slide-18

SLIDE 18

DataCamp RNA-Seq Differential Expression Analysis

Count normalization

RNA-SEQ DIFFERENTIAL EXPRESSION ANALYSIS

Mary Piper

Bioinformatics Consultant and Trainer

slide-19

SLIDE 19

DataCamp RNA-Seq Differential Expression Analysis

slide-20

SLIDE 20

DataCamp RNA-Seq Differential Expression Analysis

Count normalization

slide-21

SLIDE 21

DataCamp RNA-Seq Differential Expression Analysis

Library depth normalization

slide-22

SLIDE 22

DataCamp RNA-Seq Differential Expression Analysis

slide-23

SLIDE 23

DataCamp RNA-Seq Differential Expression Analysis

Library composition effect

slide-24

SLIDE 24

DataCamp RNA-Seq Differential Expression Analysis

DESeq2 normalization

slide-25

SLIDE 25

DataCamp RNA-Seq Differential Expression Analysis

Normalized counts: calculation

dds_wt <- estimateSizeFactors(dds_wt) sizeFactors(dds_wt)

slide-26

SLIDE 26

DataCamp RNA-Seq Differential Expression Analysis

Normalized counts: extraction

normalized_wt_counts <- counts(dds_wt, normalized=TRUE) View(normalized_wt_counts)

slide-27

SLIDE 27

DataCamp RNA-Seq Differential Expression Analysis

Let's practice!

RNA-SEQ DIFFERENTIAL EXPRESSION ANALYSIS

slide-28

SLIDE 28

DataCamp RNA-Seq Differential Expression Analysis

Unsupervised clustering analyses

RNA-SEQ DIFFERENTIAL EXPRESSION ANALYSIS

Mary Piper

Instructor

slide-29

SLIDE 29

DataCamp RNA-Seq Differential Expression Analysis

slide-30

SLIDE 30

DataCamp RNA-Seq Differential Expression Analysis

Unsupervised clustering analyses: log transformation

vsd_wt <- vst(dds_wt, blind=TRUE)

slide-31

SLIDE 31

DataCamp RNA-Seq Differential Expression Analysis

slide-32

SLIDE 32

DataCamp RNA-Seq Differential Expression Analysis

Hierarchical clustering with correlation heatmaps

# Extract the vst matrix from the object vsd_mat_wt <- assay(vsd_wt) # Compute pairwise correlation values vsd_cor_wt <- cor(vsd_mat_wt) View(vsd_cor_wt)

slide-33

SLIDE 33

DataCamp RNA-Seq Differential Expression Analysis

Hierarchical clustering with correlation heatmaps

# Load pheatmap libraries library(pheatmap) # Plot heatmap pheatmap(vsd_cor_wt, annotation = select(wt_metadata, condition))

slide-34

SLIDE 34

DataCamp RNA-Seq Differential Expression Analysis

Hierarchical clustering with correlation heatmaps

slide-35

SLIDE 35

DataCamp RNA-Seq Differential Expression Analysis

Let's practice!

RNA-SEQ DIFFERENTIAL EXPRESSION ANALYSIS

slide-36

SLIDE 36

DataCamp RNA-Seq Differential Expression Analysis

Principal Component Analysis (PCA)

RNA-SEQ DIFFERENTIAL EXPRESSION ANALYSIS

Mary Piper

Bioinformatics Consultant and Trainer

slide-37

SLIDE 37

DataCamp RNA-Seq Differential Expression Analysis

slide-38

SLIDE 38

DataCamp RNA-Seq Differential Expression Analysis

Principal Component Analysis (PCA): Theory

slide-39

SLIDE 39

DataCamp RNA-Seq Differential Expression Analysis

Principal Component Analysis (PCA): Theory

slide-40

SLIDE 40

DataCamp RNA-Seq Differential Expression Analysis

Principal Component Analysis (PCA): Theory

slide-41

SLIDE 41

DataCamp RNA-Seq Differential Expression Analysis

Principal Component Analysis (PCA): Theory

slide-42

SLIDE 42

DataCamp RNA-Seq Differential Expression Analysis

Principal Component Analysis (PCA): Theory

Sample1 PC1 score = (4 * -2) + (1 * -10) + (8 * 8) + (5 * 1) = 51 Sample1 PC2 score = (4 * 0.5) + (1 * 1) + (8 * -5) + (5 * 6) = -7 Sample2 PC1 score = (5 * -2) + (4 * -10) + (8 * 8) + (7 * 1) = 21 Sample2 PC2 score = (5 * 0.5) + (4 * 1) + (8 * -5) + (7 * 6) = 8.5

slide-43

SLIDE 43

DataCamp RNA-Seq Differential Expression Analysis

slide-44

SLIDE 44

DataCamp RNA-Seq Differential Expression Analysis

slide-45

SLIDE 45

DataCamp RNA-Seq Differential Expression Analysis

Principal Component Analysis (PCA): Theory

# Plot PCA plotPCA(vsd_wt, intgroup="condition")

slide-46

SLIDE 46

DataCamp RNA-Seq Differential Expression Analysis

Let's practice!

RNA-SEQ DIFFERENTIAL EXPRESSION ANALYSIS