Science Study | 3,000 Genes Show Sex Differences in Human Brain

A study published April 16, 2026, in Science (Vol 392, Issue 6791) has produced the most granular map of sex differences in the human brain to date. Led by Alex DeCasien, a neurogeneticist at the National Institute on Aging and the National Institute of Mental Health, the research used single-nucleus RNA sequencing (snRNA-seq) to analyze gene expression across six cortical regions in 30 adult donors, 15 male and 15 female.

The key finding is not that male and female brains are categorically different. Sex accounts for less than 1% of total variation in gene expression. But within that 1%, more than 3,000 genes showed measurably different expression levels between sexes in at least one of the six cortical regions examined, including the fusiform cortex and primary visual cortex.

While most of the 3,000 sex-biased genes were region-specific, 133 showed consistent differential expression across all brain regions and cell types examined. These are the study's most significant findings, and their chromosomal location is the most surprising result.

The dominant driver identified for autosomal gene variation was sex steroid hormones: testosterone, estrogen, and progesterone. Genes expressed more highly in females were enriched for estrogen- and progesterone-response elements. Genes expressed more highly in males were enriched for androgen (testosterone) response elements.

The study's most clinically significant finding is the direct overlap between sex-biased genes and genetic risk variants for four major psychiatric and neurological conditions. The researchers found that their identified genes frequently correlate with variants associated with diseases that already show known sex-based epidemiological patterns.

In a companion Perspective piece published alongside the study, researchers Jessica Tollkuhn and Marc Breedlove issued an important methodological caution. Because the study analyzed adult brains, it cannot currently rule out the role of socialization and lived experience.

To establish whether these 3,000 differences are present before socialization can act, future research will need to examine neonatal or fetal brain tissue. Until that data exists, the study establishes a map, not a causal chain, between biological sex, hormone signaling, and cortical gene expression. The research is foundational, but the question of how much is innate and how much is sculpted remains open.