Question

PlotCounts DESeq2 - how to plot count s for specific samples?

0

Entering edit mode

annamariabugaj ▴ 10

@e7994b78

Last seen 21 months ago

Norway

I am trying to plot counts only for wildtype and one of three mutant conditions but I don't know how to access these values form dds. My colData looks like this:

>  samples condition batch
1    C10_1       C10   b_1
2    C10_2       C10   b_2
3    C10_3       C10   b_3
4     C3_1        C3   b_1
5     C3_2        C3   b_2
6     C3_3        C3   b_3
7     C6_1        C6   b_1
8     C6_2        C6   b_2
9     C6_3        C6   b_3
10    WT_1        WT   b_1
11    WT_2        WT   b_2
12    WT_3        WT   b_3

Could someone help me on how to only plot counts for C3 samples and WT samples?

DESeq2 PlotCounts • 2.1k views

ADD COMMENT • link 21 months ago annamariabugaj ▴ 10

score 0 · Answer 1 · 2023-02-14

0

Entering edit mode

Basti ▴ 780

@7d45153c

Last seen 5 days ago

France

You can use counts(dds,normalized = TRUE) to access to normalized counts. What do you want to plot precisely ? A single gene, multiple genes ?

ADD COMMENT • link 21 months ago Basti ▴ 780

0

Entering edit mode

I woudl like to make several plots for some genes (one plot per gene) but only for samples C3_1, C3_2, C3_3 and WT_1, WT_2, WT_3. My problem is how to access the specific samples, now I am using:

    Counts <- plotCounts(dds, gene="ARHGAP21", intgroup="condition", returnData=TRUE)
colourPallette <- c("#7145cd","#bbcfc4","#90de4a","#cd46c1","#77dd8e", "#463d25")
p <- ggplot(Counts, aes(x=condition, y=count, colour = condition)) + lims(y = c(0, 2)) + geom_point() + theme_bw() + theme(axis.text.x=element_text(angle=15, hjust=1)) + scale_colour_manual(values=colourPallette) + guides(colour=guide_legend(ncol=3)) + ggtitle("ARHGAP21")
p <- p  + ylab("expression counts") 
p

And what I get is a plot for each condition (WT, C3, C6 and C10).

ADD REPLY • link 21 months ago annamariabugaj ▴ 10

0

Entering edit mode

You need to filter the samples you want to keep before creating the ggplot

ADD REPLY • link 21 months ago Basti ▴ 780

0

Entering edit mode

I am not sure how to access the samples form dds, could you please explain how to filter? edit: I found out :)

Counts <- plotCounts(dds[, c(4:6,10:12)], gene, intgroup="samples", returnData=TRUE)
colourPallette <- c("#7145cd","#bbcfc4")
p <- ggplot(Counts, aes(x=samples, y=count)) + geom_point() + lims(y = c(0, 2))  + theme_bw() + theme(axis.text.x=element_text(angle=15, hjust=1)) + scale_colour_manual(values=colourPallette) + guides(colour=guide_legend(ncol=3)) + ggtitle("gene name")
p <- p  + ylab("expression counts") 
p

ADD REPLY • link 21 months ago annamariabugaj ▴ 10

1

Entering edit mode

I would create a dds.sub object by selecting the columns of interest :

dds.sub = dds[,colData$samples[which(colData$condition %in% c("C3","WT")]]
Counts <- plotCounts(dds.sub, gene="ARHGAP21", intgroup="condition", returnData=TRUE)