I believe that they made these charts manually, using software like GraphPad Prism. I think they took the list of differentially expressed genes (up- and down-regulated separately) from their experiment and did enrichment analysis on the Gene Ontology Biological Process set using DAVID (https://david.ncifcrf.gov/summary.jsp). They just put the value of the "count" column as the x value in the graphs. I assume the * marks GO terms that were statistically significant in the input gene list.

To replicate this, you first need to generate the list of differentially expressed genes in your RNAseq data. You already have everything you need in the data you've described (gene name, padj, and fold change), you just need to apply thresholds (e.g. p < 0.05, |FC| > 2) to find the significantly differentially expressed genes. From there you can submit the gene identifiers to DAVID for analysis.

Although you won't get that exact plot, for gene ontology analysis http://www.webgestalt.org/ offers very nice tools.

The simplest thing you can do is an ORA analysis of the top differentially expressed genes.

So, you select your organism of interest, ORA as method, then the genesets database (you can start with gene ontology — biological processes non redundant). You can try many different databases and check which one works best for you.

Afterwards you will need a list of genes, usually the differentially expressed genes between two conditions, that you will paste in the corresponding field. Ideally, make two lists of genes and analyse them separately: genes with positive LFC and genes with negative LFC.

For the Reference Set, choose "genome", because if you have RNAseq analysis I assume it has genome wide data. Afterwards just click on submit and check the results.

You can also try the GSEA analysis, to do that you will need to paste into the genes field a table with two columns: genes and adjusted p-value.

Thank you very much to John and Dario for your answers.

By the moment I have found a tool called SRPlot that I didn't know and it has many options to do different bar plots (as upregulated and downregulated genes/GO terms/pathways) and other kind of plots.I will try use it to see how is.

Dario I knew Webgestalt and it is one of the tool that I use the most.

Please try muTarget

Thank you Amalesh.Is it possible with muTarget to do a bar plot of up and down regulated genes/GO terms/pathways?.I don't find how to do it with this tool.