I am a new learner of DESEQ2 and have seen similar posts on this forum but could not find clear answers.
What would be the best way to create heatmaps specifically with 2 log fold change(base 2) in DESEQ2 or R in general? In DESEQ2, heatmaps are created based on VST (Variance Stabilizing Transformation) values. The count data I've used is unnormalized raw counts from FeatureCounts.
I would like to create heatmap specifically using 2 log fold changes results from DESEQ2. I have followed the manual as standard procedure and my codes are as follows:
First, I matched row names of meta file with column names of counts file
rownames(metadata_file) <- metadata_file[,1]
all(rownames(metadata_file)== colnames(counts_file))
Then, created DESeqDataSet, subclass of RangedSummarizedExperiment
dds_file <- DESeqDataSetFromMatrix(countData=counts_file, colData=metadata_file, design= ~condition)
Results created based on alpha 0.05:
res0.05 <- results(dds_file , alpha = 0.05)
Lastly, added gene name:
res0.05$symbol <- mapIds(org, keys=row.names(res0.05), keytype="ENSEMBL", column="SYMBOL", multiVals="first")
From the full list of DESEQ2, I selected 14 genes of interests with corresponding 2 log fold changes values.
log2FoldChange
Gene_1 -1.480964026
Gene_2 -1.454365774
Gene_3 -1.842887327
Gene_4 -2.614569672
Gene_5 -0.291203886
Gene_6 -0.18059441
Gene_7 -0.499464162
Gene_8 -0.054195837
Gene_9 -5.984123536
Gene_10 -2.217040639
Gene_11 -0.146079171
Gene_12 -0.20029184
Gene_13 0.218853644
Gene_14 0.047060648
Now, I would like to create a heatmap just based on 14 selected genes and its 2 log fold changes. I read other peer reviewed articles that include heatmaps with 2 log fold changes. If anyone has suggestions, it would be greatly appreciated. Thank you.
