Demystifying the Genetics of Eating Disorders – Part II

In my last post I talked about some methods that scientists use to study the genetics of eating disorders. I focused on a subfield of genetics called behavioural genetics (which you can think of as a field that attempts to understand, in part, the interplay of genetics and environment in behaviour). In this post I’ll shift gears and focus on molecular genetics. I’ll be working of the same review paper by Drs. Zerwas and Bulik (2011). Molecular geneticists study the structure and function of genes on a molecular level. For example, they try to understand how different mutations or alterations in the genetic code can affect protein function and lead to disease states.


Genetic association studies are hypothesis driven. This means that researchers make a list of genes that are known to be involved in biological mechanisms or behaviours that are affected in eating disorders, such as appetite, mood regulation, reward, anxiety, weight regulation, and so on. Then, they compare these genes in eating disorder patients and in healthy controls to try to identify any differences between the two groups.

Most (perhaps all?) genetic association studies focused on identifying single nucleotide polymorphisms (that is, one nucleotide variations among people in the same position of the same gene, click here to find out more about SNPs) in genes associated with things like neurotransmission and hormone regulation.

The genes involved in the regulation and transmission of serotonin have received the most attention; however, there is no evidence that any of these variants confer specific risk of eating disorders. Serotonergic genes represented attractive candidates because variation in serotonin transmission and reuptake are involved in eating behavior, as well as liability to obsessive-compulsive disorder (OCD) and depressive disorders known to be comorbid with eating disorder psychopathology. […]

However, although approximately half of the studies conducted have found an association between these genes and eating disorders, the other half found no such association. This same pattern can be found in studies examining the role of other genes in eating disorders. Although the lack of replication may represent true failures to replicate, it could also reflect inadequate sample sizes and statistical power to detect replication that exists.

This is basically what I found in writing my review on the genetic and neurobiological causes of anorexia nervosa in 2008. Half the studies found an association and the other half didn’t.

Zerwas and Bulik point to small sample sizes and lack of statistical power as possible factors in explaining the contradictory results. I think it is also important to keep in mind that SNPs are not the only type of genetic variation between people. There are lots of other things we can looked at, such as copy number variants (CNVs) which have been implicated in many psychiatric disorders, including autism spectrum disorders and are really important in Huntington’s disease. (I’m not sure if there are studies on the way to examine this in eating disorder patients.)


Unlike genetic association studies, linkage analyses and genome-wide association studies (GWAS) are not hypothesis driven. This means that researchers don’t pick a list of genes they think might be associated with eating disorders or eating disorder behaviours. Instead, they survey the entire genome to find relatively large genetic regions (in linkage analyses) or sample more than half a million different SNPs across the entire genome (in GWAS) to identify if any regions or SNPs are associated with conferring risk or protection to eating disorder behaviours.

Linkage analysis studies have identified a region on chromosome one and ten that were associated with anorexia ad bulimia, respectively. These regions are associated with hundreds of genes and the next step will be to narrow down further and investigate what (if any) genes in those regions are associated with eating disorders.

When the review was written there was just one GWAS study published on anorexia nervosa. Although there were SNPs that were approaching significance, none met the threshold. As Zerwas and Bulik point out, one problem could’ve been inadequate sample size. The study had 1,033 AN patients but based on schizophrenia research, the number of patients needed is likely at least 5,000, but probably more.

It seems there was another GWAS study published in 2012 that examined disordered eating behaviours. (It is open access, so you can check it out here.) They didn’t find any SNPs that reached significance.


It is probably confusing (and even frustrating) when we contrast the difference between how much we  know about the heritability of eating disorders with how little we know about the molecular underpinnings of that heritability. One the one hand, we know genes are important–they explain 50-80% of the differences we see in eating disordered behaviours in the population–and on the hand, we know very little about what genes and gene variants are involved and how.

Why is this the case?

There are lots of reasons. Eating disorders rare enough that gathering a large enough sample size necessary to detect certain differences is a very challenging task. Secondly, researchers might be looking in the wrong places. As I mentioned earlier, SNPs are not the only type of genetic variation. There are others that (it seems to me) haven’t been investigated yet in eating disorder research. Epigenetics (which I’ll probably get to in my next post) further complicates the picture. Finally, it is very hard to study complex neuropsychiatric conditions where the same behaviour (or phenotype) can have many different genetic and neurobiological (and environmental, of course) causes. In other words, the same end state might have many different causes.

Note: I know this post was probably a bit more on the technical side. I tried to link to other resources as much as possible. Feel free to ask me anything in the comments.


Zerwas, S., & Bulik, C. (2011). Genetics and Epigenetics of Eating Disorders Psychiatric Annals, 41 (11), 532-538 DOI: 10.3928/00485713-20111017-06

Liked it? Take a second to support tet on Patreon!


Tetyana is the creator and manager of the blog.