I have result file obtained from deseq2 which has log2FC values. can anyone please tell me hoe to plot a boxplot for a gene of interest (control vs treated) from this data?
log2FC gives you the log2 Fold Change for treated relative to control.
Therefore, you will NOT be able to make two box plots: one for the treated samples and one for the control samples. This is because log2FC already describes the difference between your samples and therefore each gene will only get one log2FC value.
If you want to make two box plots, you should look at the vst function in DESeq2 which gives you the normalized RNA-seq read counts for each individual sample. You can then plot those (e.g. if you have 6 samples: 3 control samples and 3 treated samples, you can get the normalized counts for each of the 6 samples for a particular gene).
Can you be more specific about why you want to make box plots rather than just flat out reporting the log2FC value for your gene or simply displaying the log2FC value for that gene on a bar graph?
Thank you for your suggestion.
I was just trying understand how i can visualize and interpret the results and i am completely new to this kind of analysis.
Check the boxplot function and its examples. This is basic R, so please also browse the web for plotting tutorials. Have you tried anything?
yes i tried. I am dealing with cancer datasets and couldn't find anything useful which i can use for deseq2 data.
I strongly doubt that you do not find example code that plots some numeric values into boxes. What exactly is the problem?
I have DEseq2 results with log2FC and the standard error and i can plot graph using that. I just wanted to know if something else will work or not and what all are the ways to explore the data?
I recommend that the poster looks through some papers that perform RNA-seq analysis using an established pipeline like DESeq2 (there are a TON -- do a pubmed search) for ideas on how to visualize RNA-seq data, and whether your analysis is more suitable for a box plot vs. scatter plot vs. bar graph vs. heatmap vs. volcano plot (or MA plot). There are hundreds of ways to explore RNA-seq data -- it's infeasible for me to list them all here. At least describe what YOU are trying to get out of the data -- what's your experimental question?
As ATpoint mentioned, DESeq2 data just gives you numbers (log2FC's with p-values like what you have; or normalized counts). There really is nothing special you need for plotting DESeq2 data versus some other type of numerical data. If you're unsure on how to do that, I'd encourage reading more on RNA-seq because you really need to understand the data to interpret it properly.