Goheatmap

Heatmap with GO Terms

Installation

The development version can be installed from GitHub using:

devtools::install_github("ilwookkim/GOheatmap")

Usage

library(GOheatmap)

Load example data selected STAD-TCGA RNAseq data (50 patients from TP53 wildtype and 50 from mutation; 275 genes selected by top significant DEGs) mat: RSEM countdata of RNAseq, samples_df: samples annotation data frame

mat.file <- system.file("extdata", "mat.Rdata", package="GOheatmap")
load(mat.file)
ls()

[1] "mat"        "mat.file"   "samples_df"

Pre-treatment

Remove NA, if necessary.

mat <- data.frame(na.omit(mat))
knitr::kable(head(mat[, 1:4], 3), "simple")

# column: samples, row: genes (HGNC symbol)

         TCGA.CD.8536.01   TCGA.BR.8077.01   TCGA.HU.A4G3.01   TCGA.HU.A4H4.01
------  ----------------  ----------------  ----------------  ----------------
ACTC1                 31               104               349                37
ACTG2               4385              2080              6012              1236
ACTN2                  1                 6                 0                 1

Run goheatmap

goheatmap function needs the following parameters. RNAseq countdata (column: samples ID, row: HGNC gene symbol), anno (samples annotation data frame, Default to NULL) ,Parameters k (number of clustering, Default to 3), n_go (number of terms to display, Default to 3), sources details here, cor (TRUE for spearman's correlation coefficient, FALSE for vst scaled matrix, Default to TRUE), title (Title of heatmap, Default to GOheatmap)

# For the spearman's correlation coefficient matrix

goheatmap(mat, k = 3, n_go = 3, sources = "GO:BP", cor.s = TRUE, title = "GOheatmap")
goheatmap(mat, k = 3, n_go = 3, sources = "KEGG", cor.s = TRUE, title = "GOheatmap")

When sources = "", goheatmap function uses all possible sources ("GO:MF","GO:CC","GO:BP","KEGG","REAC","WP","TF","MIRNA","HPA","CORUM","HP")

goheatmap(mat, k = 3, n_go = 3, sources = "", cor.s = TRUE, title = "GOheatmap")

When cor.s = FALSE, then countdata needs to be normalized for visualization.

# For the normalized count matrix by DESeq2

if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

BiocManager::install("DESeq2")

library(DESeq2)
dds <- DESeqDataSetFromMatrix(countData = mat,
                              colData = samples_df,
                              design = ~ condition)

dds <- DESeq(dds)
mat <- as.data.frame(counts(dds, normalized=T))

If a sample annotation data frame exists, provide it into anno parameter. Default to NULL

knitr::kable(head(samples_df[, 1:2], 3), "simple")

samples           condition 
----------------  ----------
TCGA.CD.8536.01   TP53wt    
TCGA.BR.8077.01   TP53mut   
TCGA.HU.A4G3.01   TP53wt    

goheatmap(mat, anno = samples_df, k = 3, n_go = 3, sources = "GO:BP", cor.s = FALSE, title = "GOheatmap")