Volcano Plot Generator

The volcano plot exists because neither effect size nor statistical significance means much on its own. A gene with a huge fold change but a weak p-value is often a low-count artefact, and a gene with a tiny fold change but an extreme p-value may be real yet biologically trivial. Plotting log2 fold change on the horizontal axis against the negative log10 of the p-value on the vertical axis puts both judgments in a single picture.

The negative log transform of the p-value is the trick that gives the plot its shape: it stretches the vanishingly small significant values, which would otherwise pile up against zero, into a tall spray while the non-significant majority stays low. The result is a dense unchanged cloud at the base and two significant arms rising to the left and right, the down- and up-regulated genes.

Correct thresholding is what separates a defensible figure from a misleading one. Because a differential expression experiment tests thousands of genes at once, the vertical threshold should be placed on the false discovery rate, not the raw p-value; this tool applies the Benjamini-Hochberg adjustment for you. Pair it with the false discovery rate calculator to see the adjusted values directly and the log2 fold change calculator to build the horizontal axis from raw counts.

A volcano plot is the headline figure of a differential expression study, but it is the end of a pipeline that starts with normalization and a proper count model. The RNA-seq normalization calculator handles the first step, and for the complete analysis from raw counts to an annotated, corrected gene list and figure, the bioinformatics analysis service delivers reproducible results.