People with epilepsy tend to have areas of thinning (colored) in the cerebral cortex, the latest ENIGMA study has found. The darker the color, the thinner the cortex.


The world’s largest set of brain scans are helping reveal the workings of the mind and how diseases ravage the brain

ENIGMA, the world’s largest brain mapping project, was “born out of frustration,” says neuroscientist Paul Thompson of the University of Southern California in Los Angeles. In 2009, he and geneticist Nicholas Martin of the Queensland Institute of Medical Research in Brisbane, Australia, were chafing at the limits of brain imaging studies. The cost of MRI scans limited most efforts to a few dozen subjects—too few to draw robust connections about how brain structure is linked to genetic variations and disease. The answer, they realized over a meal at a Los Angeles shopping mall, was to pool images and genetic data from multiple studies across the world.

After a slow start, the consortium has brought together nearly 900 researchers across 39 countries to analyze brain scans and genetic data on more than 30,000 people. In an accelerating series of publications, ENIGMA’s crowdsourcing approach is opening windows on how genes and structure relate in the normal brain—and in disease. This week, for example, an ENIGMA study published in the journal Brain compared scans from nearly 4000 people across Europe, the Americas, Asia, and Australia to pinpoint unexpected brain abnormalities associated with common epilepsies. ENIGMA is “an outstanding effort. We should all be doing more of this,” says Mohammed Milad, a neuroscientist at the University of Illinois in Chicago who is not a member of the consortium.

ENIGMA’s founders crafted the consortium’s name—Enhancing NeuroImaging Genetics through Meta-Analysis—so that its acronym would honor U.K. mathematician Alan Turing’s code-breaking effort targeting Germany’s Enigma cipher machines during World War II. Like Turing’s project, ENIGMA aims to crack a mystery. Small brain-scanning studies of twins or close relatives done in the 2000s showed that differences in some cognitive and structural brain measures have a genetic basis. At the same time, researchers performing genome-wide association studies were scanning the DNA of large numbers of people for millions of bases that naturally vary. But efforts to link such genetic variations to differences in brain structure faltered, because imaging studies had too few participants to draw persuasive conclusions. What’s more, privacy policies often prevented researchers from sharing brain imaging data.

ENIGMA hoped to solve such problems through a meta-analysis approach, in which all the consortium’s research centers are given the same protocols to analyze brain images. Then the results are weighed and combined. Sharing processed, rather than raw, data means ENIGMA researchers don’t have to disclose details about individual patients, and the larger sample sizes enable them to draw “reliable conclusions” on how genetic variation alters brain structure, Thompson says.

In its first project, the consortium, in collaboration with another multicenter network, analyzed the brains scans and DNA of 21,000 people and linked specific genes to brain size and intelligence. For example, people with a specific variant of a gene called HMGA2 had larger brains and scored on average 1.29 points better on standardized IQ tests.

Speaking to New Scientist at the time, Steven Pinker, a neuropsychologist at Harvard University, called the result “an important finding, assuming it holds up.” But other scientists argued that the effect of HMGA2 on intelligence was too small to be significant.

Crowdsourcing brain images

Born out of a discussion during a meal in a mall, the ENIGMA brain imaging consortium has grown rapidly after a slow start.


Still, the findings, published in 2012 in Nature Genetics, gave momentum to ENIGMA’s efforts. In a follow-up study of nearly 31,000 people, the researchers examined the size of brain regions involved in memory, movement, learning, and motivation, and looked for associated genetic variants. Five newly identified genetic variants influenced the size of two regions, the putamen and caudate nucleus, the team reported in 2015 in Nature.

Such structural differences are seen not just in healthy people, but also in people with disease; both the caudate nucleus and the putamen are known to be smaller in Alzheimer’s and Parkinson’s disease patients, for example. In 2014, the ENIGMA network was awarded a 4-year, $11 million grant from the National Institutes of Health to pursue such correlations. These studies have tended to leave out genetic factors—poorly known for many diseases—and instead bring statistical power to the search for structural differences between the brains of people with and without a disorder. 

A recent example is an analysis of schizophrenia patients’ white matter, the tissue that carries impulses between neurons and across the brain. After looking at brain images from 1963 schizophrenia patients and 2359 healthy people, an ENIGMA working group reported last year in Molecular Psychiatry that the white matter wiring is impaired all over the patients’ brains, rather than only in specific areas, as many researchers previously thought. 

Also last year, ENIGMA researchers working on posttraumatic stress disorder (PTSD) found that the hippocampus—a region that is key to turning short-term memories into long-term ones—was, on average, smaller in patients than in healthy people. Some smaller studies had hinted at this reduction, which the team reported in Biological Psychiatry, but it’s now “a definitive finding,” says Rajendra Morey, a neuroscientist at Duke University in Durham, North Carolina, who leads the ENIGMA-PTSD working group.

The latest ENIGMA effort found that compared with healthy individuals, people with epilepsy tended to have areas of reduced thickness in the cerebral cortex, which influences cognition, memory, and consciousness. They also had a smaller right thalamus, a brain structure that relays information to and from the cerebral cortex. Some of these features are limited to specific types of epilepsy, the researchers found, but others appear in many epilepsies, not just the most severe. That finding, says Andrew Bagshaw, a neuroscientist at the University of Birmingham in the United Kingdom who is not part of the consortium, “starts to give a very different picture of epilepsies.”

“These changes are probably independent of the actual causes, mostly unknown,” says Sanjay Sisodiya, a neuroscientist at University College London, who led the research. The ENIGMA-epilepsy working group hopes to pursue those causes in future studies, which might track brain changes in people over time. “All ENIGMA projects start off with one relatively straightforward study and then expand as confidence grows and more people join the consortium,” Sisodiya says.

Milad considers the effort “impressive,” but says that one limitation of the study, and of most work from ENIGMA, is that the brain images come from a variety of MRI scanners and therefore might not be uniform. The challenge may grow in the future, as ENIGMA adds data on brain activity, from functional MRI or electroencephalography, to the brain mapping. By showing the brain at work, the new data should help researchers link structural differences to function, but they could prove even more difficult to standardize and weigh, Milad says.

Emily Jones, a neuroscientist at Birkbeck University in London, echoes this concern but adds that the large sample sizes available to ENIGMA researchers should help compensate for inconsistencies between labs. 

Geneticist and ENIGMA Co-Founder Barbara Franke of Radboud University in Nijmegen, the Netherlands, expects that as the consortium combines genetics with data from ever larger numbers of brains, it will start to trace the pathways leading from genes to brain alterations to diseases. Ultimately, diagnosis and prevention will benefit, Franke predicts. However, she cautions, “We’re still a long way from it.”

*Correction, 24 January, 10 a.m.: An earlier version of this story misstated the journal that published the first ENIGMA study.