Plotting TPM counts for one crab over 3 days

[aspencoyle]

Sorry for the delay - here's all TPM counts for a single crab (specifically Crab A) over time. This particular crab was kept at ambient temperature throughout the course of the experiment

For background, this library was constructed by aligning the library to cbai_transcriptome_v4.0, which was constructed by BLASTing against the C. opilio genome. In other words, this library should contain only C. bairdi sequences.

Crab_A_cbai_transcriptome_v4.0.txt

[sr320]

Looking for anyone to plot these data and post here

[laurahspencer]

Genes have been filtered to retain those with 5+ reads across all 3 time points

Code:
pheatamp_cbairdi.zip

[sr320]

I like it! to me there are clear clusters!

[sr320]

@laurahspencer is there a way to pull genelists from this based on clustering?

[kubu4]

@grace-ac You did something like this with the crab clustering, right? Can you point to some code to pull info from clusters?

[grace-ac]

Yep! here's a script: https://github.com/RobertsLab/paper-tanner-crab/blob/master/scripts/heatmaps-DEGs_in-singlecrab_over-time.Rmd

[laurahspencer]

Yes the script in my post should let you pull gene lists from clusters - I didn’t dive too deep into it, but you define the number of clusters you want, then extract gene lists from there

…

On Tue, May 11, 2021 at 11:41 AM grace-ac ***@***.***> wrote: Yep! here's a script: https://github.com/RobertsLab/paper-tanner-crab/blob/master/scripts/heatmaps-DEGs_in-singlecrab_over-time.Rmd — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1206 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEDXA7JC3MB66NOZ4XYAYETTNF25HANCNFSM44ULNIFA> .

[laurahspencer]

Apparently the script I included in that zip folder was the wrong one! Here's the code:

Load libraries and source scripts

list.of.packages <- c("pheatmap", "tidyverse") #add new libraries here 
new.packages <- list.of.packages[!(list.of.packages %in% installed.packages()[,"Package"])]
if(length(new.packages)) install.packages(new.packages)

# Load all libraries 
lapply(list.of.packages, FUN = function(X) {
  do.call("require", list(X)) 
})

Load count data

(data <- read.table(file = "Crab_A_cbai_transcriptome_v4.0.txt"))

# Drop any genes with less than 5 counts across all 3 samples 
keep <- rowSums(data) >= 5
data.f <- data[keep,]
print(paste("# genes remaining after pre-filtering:", nrow(data.f)))
print(paste("# of genes dropped:", nrow(data) - nrow(data.f), sep=" "))

Construct heat map

pheatmap(data.f, cluster_rows=TRUE, show_rownames=F, na.rm=TRUE, cluster_cols = FALSE, scale="row", main = "gene counts by day", fontsize = 9) 

# extract gene clusters 
out <- pheatmap(data.f, cluster_rows=TRUE, show_rownames=F, na.rm=TRUE, cluster_cols = FALSE, scale="row", main = "gene counts by day", fontsize = 9) 
hc <-out$tree_row
lbl <- cutree(hc, 4) # split gene dendrogram in 4 groups
which(lbl==1) %>% names() # grab genes of first group

Extract counts for the 4 gene clusters

data.f[which(lbl==1) %>% names(),]
data.f[which(lbl==2) %>% names(),]
data.f[which(lbl==3) %>% names(),]
data.f[which(lbl==4) %>% names(),]

[laurahspencer]

You can split the dendogram into more/less clusters above at the cutree function.
Also, you can generate heat maps for gene sets in each cluster, like this:

clust1 <- data.f[which(lbl==1) %>% names(),]
clust2 <- data.f[which(lbl==2) %>% names(),]
clust3 <- data.fwhich(lbl==3) %>% names(),]
clust4 <- data.f[which(lbl==4) %>% names(),]

pheatmap(clust1, cluster_rows=TRUE, show_rownames=F, na.rm=TRUE, 
cluster_cols = FALSE, scale="row", main = "gene counts by day, cluster 1", fontsize = 9)