impulseDE2 result table

Question

[Josephine]

Hi biostars,

I'm using impulseDE2 to do differential expression analysis of time course RNA-seq data, and there're some details I can't understand in the result table. My data contains 5 time points and 53 samples from 4 batches.

Q1: If I set boolIdentifyTransients = TRUE, I get some genes (padj < 0.05) with "is.transient=FALSE" "is.monotonous=FALSE". What are these genes? How do they change over time?

Q2: When plotting heatmap with the built-in function, it seems like the gene number of "boolIdentifyTransients = TRUE" and "boolIdentifyTransients = FALSE" is not same. And how do I get the genelist of high z-score genes(which are red in the heatmap)?

Any ideas will be greatly appreciate! Thanks!

Here is my code: (sorry for the poor format)

result <- runImpulseDE2(
  matCountData = readscount,
  dfAnnotation = group,
  boolCaseCtrl = FALSE,
  vecConfounders = c("Batch"),
  scaNProc = 4,
  scaQThres = 0.05,
  boolIdentifyTransients = TRUE
)

library(ComplexHeatmap)

lsHeatmaps3 <- plotHeatmap(
  objectImpulseDE2       = result,
  strCondition           = "case",
  boolIdentifyTransients = TRUE,
  scaQThres              = 0.01)
draw(lsHeatmaps3$complexHeatmapRaw)

[cat.lou]

Hi Josephine, did you ever figure out the answer to Q1? I'm also trying to figure out the reason for this.

[kelen]

For Q1. If it helps, he has addressed the same question on Github - https://github.com/YosefLab/ImpulseDE2/issues/5

Q2. If I remember correctly I never figured out an easy way to do this, but if you check the source code for the heatmap (https://github.com/YosefLab/ImpulseDE2/blob/master/R/srcImpulseDE2_plotHeatmap.R) you can trace back what is plotted and how to retrieve it.