When performing differential expression analysis, why is it necessary to apply the Benjamini-Hochberg procedure instead of a standard p-value threshold across thousands of genes?

When performing differential expression analysis, you test thousands of genes simultaneously, which creates a major statistical problem known as the multiple testing problem. A standard p-value represents the probability of observing a result by pure chance if there is actually no true difference between conditions. If you set a common threshold like 0.05, you are stating that there is a five percent risk of a false positive for each individual test. When you perfor....