This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Worrisome dispersion plot interpretation (DESeq2)

Hi, I need some help interpreting a dispersion plot—specifically, what the estimated genes in the top‑right corner represent. Are those points outliers? Also, what exactly is happening with the data since the black dots aren’t around the fitting curve? The left image is identical to the right one; I only changed the 'final' and 'fitted' to white colour so the gene‑level estimates would stand out more clearly. thank you

enter image description here both axis in log scale

plot deseq2 deseq2 dispersion
• 663 views • link
written by IrK IrK

What are the data, are these raw counts? Is this RNA-seq? Show code.

• 0 views • link
written by ATpoint ATpoint

RNA-seq - 5 genotypes, two time points (early vs late), condition (ctrl vs treat). 3 replicates for each condition

dds <- DESeqDataSetFromMatrix(countData = count_data_mtx,
                              colData   = meta_data,
                              design    = ~ condition+time+genotype) 
dds

# Pre-filtering
keep <- rowSums(counts(dds) >= 10) >= smallestGroupSize
dds <- dds[keep,]
# Add a pseudo-count value of '1' to data
assays(dds)[["counts"]] <- round(as.matrix(assays(dds)[["counts"]]))+1
storage.mode(assays(dds)[["counts"]]) <- "integer"

dds <- DESeq(dds, fitType='local')

plotDispEsts(dds, genecol ="black", fitcol = "white", finalcol = "white");
plotDispEsts(dds, genecol ="black", fitcol = "orange", finalcol = "grey");
• 0 views • link
written by IrK IrK

0 answers

No answers yet.

Log in to answer this question.

More posts like this