Hope this helps
https://stackoverflow.com/questions/52136211/how-to-reorder-cells-in-doheatmap-plot-in-seurat-ggplot2

Re point 2, DoHeatmap doesn't support these kinds of dendrogram plots. However, it's worth noting that in order to draw those dendrograms, the cells would still need to be clustered in some way so I'd just recommend re-clustering the new object in that scenario.

Hi @summerrfair ,
could you manage to do it?
the link provided by @mxposed doesn't help me ...

I also have a DoHeatmap that I want to order from low to high expression on 1 gene (multiple genes on my heatmap, but we care about one in particular)

any help would be greatly appreciated :)

not very nice, and not particularly efficient but something like this might work and might help you get to where you want to be.

extract data for gene of interest
x <- temp.seurat@assays$originalexp@data["Ifng", ]

x <- t(x)
x <- as.data.frame(x)
order.index <- order(x$V1, decreasing = T)
cell.order <- rownames(x)[order.index]

DoHeatmap(temp.seurat,
cells = cell.order,
features = c("Ifng", "Tnf"))

not very nice, and not particularly efficient but something like this might work and might help you get to where you want to be.

extract data for gene of interest x <- temp.seurat@assays$originalexp@data["Ifng", ]

x <- t(x) x <- as.data.frame(x) order.index <- order(x$V1, decreasing = T) cell.order <- rownames(x)[order.index]

DoHeatmap(temp.seurat, cells = cell.order, features = c("Ifng", "Tnf"))

thanks @Eomesodermin ,

the 1st line was not working for me, so I did:

x<- FetchData(SAMPLE, vars = "Gene", use.scaled = TRUE) 
x <- as.data.frame(x)
head(x)
order.index <- order(x$Gene, decreasing = FALSE)
cell.order <- rownames(x)[order.index]
head(cell.order)
  
  DoHeatmap(object = SAMPLE, cells = cell.order,features=c(Gene,"xxx","yyy","zzz"), group.by = "Pain_score", label = TRUE, angle=45) 

and it worked :)
thanks!