What Happens in the Brain in Mental Disorders? A New Map of the Human Prefrontal Cortex

On September 23, 2026, a series of papers from the PsychAD project was published in Nature. One of its most important outcomes is an exceptionally detailed molecular atlas of the dorsolateral prefrontal cortex, or DLPFC. This region of the brain is involved in working memory, concentration, planning, decision-making, and behavioral control. Dysfunction of the networks associated with this region has long been studied in schizophrenia, bipolar disorder, and other brain disorders.

The scale of the project is impressive. Researchers used brain tissue from nearly 1,500 donors and analyzed more than 6.3 million individual cell nuclei. The study population included people with schizophrenia, bipolar disorder, and neurodegenerative diseases. One important limitation should be emphasized: the main atlas did not include a comparably large group of patients with unipolar depression, so findings concerning schizophrenia and bipolar disorder cannot automatically be extrapolated to depression.

Perhaps the most interesting aspect, however, is the method itself. Until relatively recently, analyzing a piece of brain tissue was somewhat like trying to understand conversations taking place in a stadium by measuring the average noise level. It was clear that something was happening, but it was difficult to determine who was actually speaking. The brain is not composed of a single type of cell. Alongside different populations of neurons, there are astrocytes, oligodendrocytes, microglia, and cells associated with blood vessels. Each performs different functions.

Single-nucleus RNA sequencing makes it possible to separate this biological crowd. Researchers can determine which genes are active in specific types of cells. In one of the atlases, researchers identified 8 major classes, 27 subclasses, and 65 more specific cell types within the prefrontal cortex. This allows the question “What changes in the brain in mental illness?” to be replaced by a much more precise one: “In which cells does a particular change occur?”

This has major implications for psychiatric genetics. A gene statistically associated with the risk of a disorder does not necessarily act in the same way throughout the brain. Its significance may become apparent primarily within a particular population of neurons or glial cells. Moreover, many DNA variants associated with mental disorders are located outside regions that directly encode proteins. They may influence gene regulation, for example by determining when, where, and how strongly a particular gene is activated. The new atlases help connect these elements and identify the cells in which genetic risk may have biological significance.

This also helps explain why researchers have not found a single “schizophrenia gene” or “bipolar disorder gene.” Both disorders have complex, polygenic origins. Risk is influenced by many genetic variants, each of which typically contributes only a small proportion of the overall risk. Genetics also does not operate independently of brain development, age, and environmental influences.

The new research highlights another issue: the boundaries between mental disorders are not always as distinct at the biological level as diagnostic labels might suggest. Researchers identified both changes characteristic of particular disorders and molecular programs occurring across several different conditions. This does not mean that schizophrenia and bipolar disorder are the same disease. It indicates that some biological mechanisms may be shared, while others remain specific to a particular disorder.

A similar situation can be observed in psychiatric practice. Problems with concentration may occur in depression, ADHD, schizophrenia, and bipolar disorder. Psychotic symptoms may occur in schizophrenia, severe depressive episodes, or mania. Sleep disturbances are present in many mental disorders. A similar symptom therefore does not necessarily indicate an identical biological mechanism.

This is where the limitations of the long-standing popular explanation that depression results from a “serotonin deficiency” and schizophrenia from an “excess of dopamine” become particularly apparent. Neurotransmitters remain extremely important and are the targets of many effective medications, but mental disorders cannot be reduced to a simple deficiency or excess of a single substance.

SSRIs provide a good example. Their effect on the serotonin transporter begins relatively quickly, while the antidepressant effect usually develops much more slowly. This means that a range of adaptive processes occurs between the initial pharmacological action of the drug and clinical improvement. Current research examines, among other things, changes in receptor function, synaptic plasticity, the activity of entire neuronal networks, gene expression, and interactions between neurons and glial cells. Serotonin is therefore one part of a much larger biological puzzle.

The situation is similar with dopamine and schizophrenia. The effectiveness of medications acting on dopamine receptors is an important part of our understanding of the biology of psychosis, but it does not demonstrate that schizophrenia is simply a disease caused by “too much dopamine in the brain.” Specific dopaminergic pathways, other neurotransmitter systems, the development of neuronal networks, genetic factors, and processes occurring at the synaptic and cellular levels all play a role.

Another part of the project provided interesting information about how the healthy prefrontal cortex changes with age. Researchers analyzed more than 1.3 million cell nuclei from 284 individuals spanning virtually the entire human lifespan. Gene activity in the brain did not change in a simple, linear fashion. Researchers observed a period of intensive remodeling during development, relative stability during adulthood, and another wave of changes later in life.

The period after approximately age 65 proved particularly interesting. Genetic programs associated with glial cell activity, immune responses, and cellular stress became increasingly prominent. This does not mean that the brain suddenly begins to develop disease after a certain age. It does, however, show that the biological environment in which neurons function changes with aging.

The biological clock provides an even more intriguing aspect of the findings. Researchers observed age-related reorganization of the expression of genes involved in circadian rhythms. In neurons, some molecular rhythms became less synchronized, while rhythmic programs associated with processes such as cellular stress emerged in glial cells. This does not mean that researchers have discovered a single cause of insomnia or depression in older adults. It does show, however, that circadian rhythms can also be observed at the level of gene activity in specific brain cells.

Are we therefore approaching a point at which depression, schizophrenia, or bipolar disorder will be visible on magnetic resonance imaging? For now, the answer is no. Neuroimaging studies can identify statistical differences between large groups of patients and healthy individuals, but these differences are not sufficiently specific to diagnose most mental disorders in an individual patient. The brains of healthy individuals also vary considerably, and findings from different groups often overlap.

For this reason, a normal MRI scan does not rule out depression, schizophrenia, or bipolar disorder. At the same time, identifying a particular structural difference in the brain does not automatically establish any of these diagnoses. MRI can, however, be extremely important when a physician needs to exclude a neurological or other organic cause of a patient’s symptoms.

The new cellular maps should be interpreted in the same way. They are powerful scientific tools, but they are not diagnostic tests. A blood sample cannot be taken from a patient and used together with the PsychAD atlas to confirm schizophrenia or bipolar disorder. Nor is there a genetic test that can determine with certainty whether a healthy person will develop one of these disorders in the future.

One of the most interesting findings of the project is the enormous degree of variation between individuals. Even within the same cell population, gene activity may differ substantially from one person to another. This may be one reason why identifying a single psychiatric biomarker has proved so difficult.

At the same time, this diversity may eventually become a major opportunity. If researchers learn to identify biological subtypes of mental disorders, psychiatry may gradually move toward more personalized treatment. Two patients who meet the criteria for the same diagnosis could receive different treatments, not only because they previously responded differently to medications, but because their disorders involve different biological mechanisms.

For now, this remains a direction of research rather than a standard part of everyday clinical practice. Discovering a gene, cell type, or molecular pathway associated with a disorder does not automatically mean that a medication exists that can safely modify that mechanism. Years of laboratory and clinical research often separate the discovery of a potential biological target from the development of an effective treatment.

The new brain maps are, however, changing the way mental disorders can be understood. The brain of a person with bipolar disorder, schizophrenia, or depression is not a device in which one simply needs to identify a substance that is present in insufficient amounts and replace it. It is an extraordinarily complex system composed of different types of neurons and glial cells, millions of connections, and thousands of active genes whose activity changes with development, aging, and environmental influences.

The PsychAD atlas does not yet provide psychiatrists with a new test that can be ordered during a clinical consultation. It does, however, provide researchers with something equally important: a far more detailed map of the biological landscape in which future biomarkers and new therapeutic targets will be sought.

References: