code I could use as a template for heatmap dot 2 for my own data 11-13-2014d1019

2015-04-13

code I could use as a template for heatmap dot 2 for my own data 11-13-2014d1019
code modified from here http://www.molecularecologist.com/2013/08/making-heatmaps-with-r-for-microbiome-analysis/


Note the last time I tried to run this heatmap code, I received the following error: 04-13-2015d1750 


Error in heatmap.2(as.matrix(all.data), Rowv = as.dendrogram(row.clus),  : 


  object 'var1' not found


Does this code cluster both rows and columns?
Yes


What should the form of the input data look like? 





    sample Acetivibrio Acetobacter Achromobacter





1     S7           0           5             0





2     S8           1           0             1





3     S9           0           3             1



What should be changed in the code for it to work with your specific data? 


Note that the data should be in csv format





change the line with


row.names(all.data) <- all.data$sample


so that "sample" is replaced with the title of your different rows.








change the line with


all.data <- read.csv("C:/Sync/pDR/2014/2014/06-26-2014d0006/test_file4.csv")  # load the data


so that it reads your data file



code to make heatmap in R

library(gplots) # for heatmap.2
# to install packages from Bioconductor:
source("http://bioconductor.org/biocLite.R")
biocLite("Heatplus") # annHeatmap or annHeatmap2
library(Heatplus)

# load the vegan package for hierachical clustering if you want to use distance functions not specified in dist.
library(vegan)

# load the RColorBrewer package for better colour options
library(RColorBrewer)

all.data <- read.csv("C:/Sync/pDR/2014/2014/06-26-2014d0006/test_file4.csv") # load the data
dim(all.data)

row.names(all.data) <- all.data$sample
all.data <- all.data, -1

# colorRampPalette is in the RColorBrewer package. This creates a colour palette that shades from light yellow to red in RGB space with 100 unique colours
scaleyellowred <- colorRampPalette(c("lightyellow", "red"), space = "rgb")(100)

# calculate the Bray-Curtis dissimilarity matrix on the full dataset:
data.dist <- vegdist(all.data, method = "bray")

# Do average linkage hierarchical clustering. Other options are 'complete' or 'single'. You'll need to choose the one that best fits the needs of your situation and your data.
row.clus <- hclust(data.dist, "aver")


# you have to transpose the dataset to get the genera as rows
data.dist.g <- vegdist(t(all.data), method = "bray")
col.clus <- hclust(data.dist.g, "aver")

# make the heatmap with Rowv = as.dendrogram(row.clus)
heatmap(as.matrix(all.data), Rowv = as.dendrogram(row.clus), Colv = as.dendrogram(col.clus), col = scaleyellowred, margins = c(10, 3))




heatmap.2(as.matrix(all.data), Rowv = as.dendrogram(row.clus), Colv = as.dendrogram(col.clus), col = scaleyellowred, RowSideColors = var1, margins = c(11, 5), trace = "none", density.info = "none", xlab = "genera", ylab = "Samples", main = "Heatmap example", lhei = c(2, 8)) # this makes the colour-key legend a little thinner