Share this article on:

What to Make of the Daily Published Genomic Associations?: Neurogeneticists Urged to Probe Deeper


doi: 10.1097/01.NT.0000360715.31197.b6

ARTICLE IN BRIEF Experts in neurogenetics wade through the welter of confusing and contradictory findings in the field to discern what is translatable into new treatments or prognostic tests.



Nine years after the completion of a rough draft of the human genome, neurologists and geneticists who hoped the so-called “book of life” would prove to be easier to read than a Harlequin romance have instead concluded that it's more on the order of James Joyce's Ulysses: as opaque as it is protracted.

But, they insist, patience in awaiting clinically useful findings will be rewarded.

“The whole field is less than ten years old — we're just beginning,” said Roger N. Rosenberg, MD, director of the Alzheimer's Disease Center at the University of Texas Southwestern Medical Center in Dallas and editor of a major textbook on neurogenetics. “Clinicians should not lose heart. The human genome was only sequenced in 2000. We've got the dictionary; now we've got to learn how to read it.”

Still, the welter of confusing and contradictory findings in the field of neurogenetics has been frustrating for neurologists who believed that promising early reports would quickly translate into new treatments or prognostic tests.

“Most of the early genetic associations reported in neurology, whether for predisposition or treatment response, turned out to be just not true when examined in subsequent studies,” said David B. Goldstein, PhD, director of Duke University's Center for Population Genomics and Pharmacogenetics.

Back to Top | Article Outline


The latest fireball to crash and burn in the field of neurogenetics, or at least in the closely related field of psychiatric genetics, was a much ballyhooed finding, published six years ago in Science, that allegedly established a link between genes, life events, and depression. The 2003 study followed 847 people from birth to age 26, and compared their rates of depression based on two variables: whether or not they experienced stressful experiences such as divorce, abuse, or unemployment; and whether they had zero, one or two copies of a shortened version of the 5-hydroxy-tryptamine (5-HTT) or serotonin transporter gene, which encodes the protein that transports serotonin across neuronal synapses.

The 5-HTT gene had been targeted because serotonin had long been implicated in the risk of depression, particularly because drugs known as “selective serotonin reuptake inhibitors (SSRIs)” have proved to be moderately effective in treating the disorder.



Among those who experienced stressful life events between the ages of 21 and 26, according to the 2003 study, 43 percent of people with two copies of the shorter version of the gene developed depression, compared to 33 percent who had one copy of the shorter version, and 17 percent of those with zero copies.

But because subsequent studies reached conflicting conclusions on the role of this gene, a team of researchers coordinated by the National Institute of Mental Health performed a meta-analysis of 14 prior papers, including 10 for which the original data was re-analyzed to assure comparability. Their conclusion, published on June 17 in JAMA, was that the original finding simply did not hold up.



“It should make people think twice about these candidate gene studies,” said the first author of the study,” said Neil Risch, PhD, distinguished professor of human genetics and director of the Institute for Human Genetics at the University of California-San Francisco.

Back to Top | Article Outline


So-called “candidate” gene studies were based on logical assumptions about genes thought likely to cause neurologic diseases. Hundreds of studies conducted early in the decade came up with apparent hits, the vast majority of which did not hold up in subsequent efforts to validate the initial findings.

Then came the genome-wide association studies, using gene chips with 500,000 to even a million single-nucleotide polymorphisms, or SNPs (pronounced ”snips,”) have found dozens of areas associated with disease. The value of these studies, Dr. Risch said, is clear: “You don't start with a prejudice. Most of the genes that have come out from linkage studies and genome-wide association studies have been surprises.”

Even many of the genome-wide association studies, however, have reached conflicting conclusions.

One area of neurology in which genome-wide association studies have proved especially fruitful, Dr. Risch said, is in multiple sclerosis. He cited, for instance, a paper published in July in Nature Genetics, in which a meta-analysis of prior studies involving 2,624 MS subjects and 7,220 controls was combined with a new analysis of another 2,215 subjects and 2,116 controls to identify three new susceptibility loci.

Many other genome-wide association studies, however, have reached conflicting conclusions. “Suffice it to say that the strength of the epilepsy association studies published so far is moderate or weak,” said Nigel C.K. Tan, MRCP, a researcher at the National Neuroscience Institute in Singapore.

The International League Against Epilepsy's genetics commission has just completed a report on genetic testing, said its immediate past chair, Ruth Ottman, PhD, professor of epidemiology (in neurology and the Sergievsky Center) at Columbia University.

“We concluded there are few instances in which it would make sense to do predictive testing in family members,” Dr. Ottman said. “In diagnostic testing, only one example is clear: Dravet syndrome (or severe myoclonic epilepsy of infancy). More than 70 percent of cases have a mutation in the sodium channel voltage-gated, type I alpha (SCN1A) gene and testing helps to confirm the diagnosis.”

To help clinicians and researchers keep track of the shifting fortunes of genes associated with epilepsy, Dr. Tan co-founded the Epilepsy Genetic Association Database (epiGAD), at

Neurologists interested in Parkinson disease, meanwhile, can keep up with the latest genetic findings at And those who specialize in Alzheimer disease (AD) can stay up to date at, developed by Lars Bertram, MD, head of the Neuropsychiatric Genetics Group at the Max-Planck Institute for Molecular Genetics in Berlin, Germany.

“We've seen hundreds of times in Alzheimer disease that genes considered important in one study were not supported by other studies,” Dr. Bertram said. “That's why we developed the AlzGene site, to know which are the most promising.”

One of the few genes to hold up repeatedly and to show a significant effect on a common neurologic disease is the apolipoprotein-E4 (APOEe4) allele for AD, Dr. Bertram said. “APOEe4 has been replicated unequivocally,” he said. “It was a low-hanging fruit. Now we're after the higher hanging fruit which is more difficult to find.”

Back to Top | Article Outline


But a critical problem now looms before researchers: aside from a few welcomed exceptions like APOEe4, the vast majority of the remaining associations that have been validated carry only a slim increase in disease risk, rarely more than a few percentage points.

“They have tiny, tiny effects,” said Dr. Goldstein, who published an influential paper on the subject in April in the New England Journal of Medicine. “The associations are real, but do they matter? That's what the current debate is about.”

While some argue that the field is now on the right track and should continue conducting genome-wide association studies with gene chips containing 500,000 or more SNPs, Dr. Goldstein says a new approach is necessary. “I do not think that we need to keep doing studies collecting those tiny effects,” he told Neurology Today. “I don't think that's useful in terms of diagnostics, in terms of predicting individual's risks.”

Rather, he and others insist, the field must embrace a costly new strategy: analyzing all 3 billion base pairs of the genome, searching not just the protein-coding sections of genes but the whole shebang, in the hopes of finding administrative regions that control how many of the proteins will actually get transcribed.



“A number of genetic researchers, including myself, would say the era of searching SNPs has been necessary but may not be sufficient to find the hidden heritability for Alzheimer disease and other neurological disorders,” Dr. Rosenberg said. “The whole genome is going to have to be sequenced, not just the 500,000 or so SNPs that have been investigated so far.”

Dr. Bertram agreed. “Clearly that's where the field is going,” he said. But, he added, “Even if you have the complete sequence, if you know every single base pair in an individual, then you need to understand it — and that's where the problem lies. That will keep geneticists and bioinformaticians pretty busy for years to come.”

A possible outcome of whole-genome studies, Dr. Goldstein said, could be that most of the common diseases of neurology are caused by any of a large number of mutations, each of which is rare, but any of which carries a devastatingly large burden of risk.

“The hope for personalized medicine has been that maybe there are four or five underlying genetic causes of, say, Alzheimer disease,” he said. While his view of many potential causes might seem to dash that hope, he offered a way out: “The light at the end of the tunnel for the rare variant perspective is that all or most of these different genetic causes might still converge on only a few key pathways.” Still, he said, “Right now it's too early to tell if that will be the case.”

Dr. Risch, however, believes that the so-called “rare variant, large effect” model is unlikely to hold up. “A more likely scenario,” he said, “is lots of low frequency variants of moderate to small effect. Those are even harder to find statistically than the common variants (even doing sequencing), because the statistical power is low when the allele frequencies are small, and it will also be complicated by the fact that there will probably be millions of low frequency alleles to test rather than 500,000.”

Whichever explanation is ultimately validated, Dr. Rosenberg emphasized his conviction that there will be a happy ending to this saga.

“These loci that show variation that will be found by sequencing, and will become potential molecular targets for new drug discovery,” he said. “This is just the beginning of the ballgame.”

In the meanwhile, neurologists seeking help in making heads or tails of the latest studies are encouraged to read the full text of three papers published on Jan. 9 in JAMA, which together offer a thorough overview of the subject.

Back to Top | Article Outline


• Caspi A, Sugden K, Moffitt TE, et al. Influence of life stress on depression: moderation by a polymorphism in the 5-HTT gene. Science 2003; 301 (5631): 386–389.
    • Risch N, Herrell R, Lehner R, et al. Interaction Between the serotonin transporter gene (5-HTTLPR), stressful life events, and risk of depression: A meta-analysis. JAMA 2009; 301 (23): 2462–2471.
      • De Jager PL, Jia X, Wang J. Meta-analysis of genome scans and replication identify CD6, IRF8 and TNFRSF1A as new multiple sclerosis susceptibility loci. Nat Genet 2009;41(7):776–782. E-pub 2009 Jun 14.
        • Waring SC and Rosenberg RN. Genome-wide association studies in Alzheimer disease. Arch Neurol 2008; 65:329–334.
          • Goldstein DB, Common genetic variation and human traits. N Engl J Med 2009; 360 (17): 1696–1698.
            • Attia J, Ioannidis JPA, Thakkinstian A, et al. How to use an article about genetic association. JAMA 2009; 301 (1):74–81.
              ©2009 American Academy of Neurology