Mention “genes” and “race” in the same breath, and someone is likely to take offense. But while research on genetic differences that neatly track the three generally-recognized races has mostly come up empty—non-overlapping genetic differences track ethnic or population groups better than they do races, which are too large to be biologically meaningful—a new approach may yield more.

Or so believe scientists who will report next week that people of African and European ancestry differ in the levels of gene expression for 383 out of the 9,156 human genes they examined.

People are more than 99.9 percent identical at the level of their DNA, the Human Genome Project has found. But which genes you have isn’t the whole story. What also matters is how strongly those genes are expressed, or turned on. To use an analogy I’ve tried before, DNA is like your iPod downloads: what counts isn’t only what you’ve stored, but whether you play it. So what matters about DNA is not only which genes you have, but which ones are expressed and at what levels.

The finding that expression levels differ for hundreds of genes between people of African and European ancestry (the scientists studied 30 white families from Utah and 30 Yoruban families from Nigeria) links up with earlier observations. The researchers found differences among genes that help produce the antibodies that fight off bacterial invaders, and among genes that influence how you respond to certain drugs, they write in the American Journal of Human Genetics. Not coincidentally, perhaps, a 1980 study found that African Americans may be more susceptible to certain bacterial infections than are people of European ancestry. And a 2005 study found that individuals of European and African ancestry also differ in how they respond to anti-microbial drugs such as those used to treat gum disease.

The hope is that understanding the genetic basis for different responses to drugs could help direct treatment—the great hope (but still only a hope, for the most part) of personalized medicine.

But this work is interesting for another reason. It emphasizes again that what counts in genetics is not merely having a gene but expressing it. If your iPod never plays that Amy Winehouse you downloaded, then (in a tree-falling-in-an-empty-forest sense) you do not have the gene.

Patterns of expression—on in this person, off in that—likely lie behind such mysteries as why identical twins are not identical, especially in which supposedly “genetic” diseases they get. This field is called epigenetics. The Human Genome Project was all about determining the sequence of the 6 billion building blocks that constitute human DNA. But that sequence is only the start.