An Evolutionary Perspective on the Genetics of Anorexia Nervosa
===============================================================

* Édith Breton
* Tobias Kaufmann

## Abstract

Anorexia nervosa (AN) is a serious psychiatric disorder that typically emerges around adolescence and predominantly affects girls and women. Recent progress has been made in identifying biological correlates of this debilitating disorder, but more research is needed to pinpoint the specific mechanisms that lead to the development and maintenance of AN. There is a known phenotypic link between AN and growth as well as sexual maturation, yet the genetic overlap between these phenotypes has never been investigated in detail. One may hypothesize that shared factors between AN, energy metabolism and reproductive functions may have been under recent evolutionary selection. Here, we characterize the genetic overlap between AN, BMI and age at menarche, and took an evolutionary perspective, aiming to reveal potential evolutionary factors that may help explain the origin of AN in recent human evolution. We obtained publicly available GWAS summary statistics of AN, BMI and age at menarche and studied the polygenic overlap between them. We found 22 loci overlapping between AN and BMI, and 9 loci between AN and age at menarche, with twelve of these not previously associated with AN. Next, we used Neandertal Selective Sweep scores to explore potential positive evolutionary selection for loci associated with the three phenotypes, or overlapping between them. We found evidence that gene variants associated with AN may have been under particular evolutionary dynamic. The circadian cycle seemed to be highly relevant to the studied genetic overlaps and prone to recent evolutionary selection, offering a promising avenue for future research into chronobiology. Taken together, our findings contribute to the understanding of the genetic underpinning of AN. Ultimately, better knowledge of the biological origins of AN may help to target specific biological processes and facilitate early intervention in individuals who are most at risk.

Key words
*   Anorexia Nervosa
*   Human Evolution
*   Genetic Overlap
*   Age at Menarche
*   Body Mass Index
*   Eating Disorders

## Introduction

Anorexia Nervosa (AN) is a serious psychiatric disorder with long-lasting consequences for physical health, psychological well-being and quality of life (1–3). Symptoms of AN typically include severe restriction of energy intake, intense fear of gaining weight, and disturbed body image (2). Notably, research on AN and other eating disorders (ED) has lagged significantly behind that of other major psychiatric disorders (2,4), particularly with regard to the biological mechanisms involved in their development. Yet, AN has one of the highest mortality and morbidity rates of all psychiatric disorders – with a standardized mortality ratio of 5.9 (5).

Restrictive EDs affect approximately one male for every 10 females, and there is evidence to suggest that puberty and sex hormones may influence these discrepancies across sexes. Relatedly, EDs commonly emerge around puberty, with a median age at onset of 12 years (6). Major neurodevelopmental changes occur during this period, potentially increasing the vulnerability to develop a mental disorder (7). Growth may also play a role, since children with both lower and higher body mass index (BMI) before adolescence may be at increased risk of EDs (8–11). Significant body changes and increased peer influence during puberty may be additional triggering factors for some individuals (12,13).

Earlier age at onset and time spent with an ED are predictors of poorer outcomes, supporting the need to intervene in the early stages of the illness (2,3,14,15). Evidence from longitudinal studies suggests that ∼30% of individuals with AN will recover in the first 10 years after diagnosis, and an additional ∼30% will recover after 20 years, sometimes without any treatment (1,16,17). Many people with AN will experience relapses (1,3) or will continue to experience some psychiatric symptoms after remission from their ED (16,18). Overall, this suggests fair chances for recovery for individuals with AN, yet that a third of them will still be ill after 20 years and that a large proportion will live with psychiatric symptoms for a significant part of their lives (1).

Understanding the genetic underpinnings of AN could provide an opportunity to identify biological mechanisms at play – which could eventually become targets for improving both pharmacological and psychosocial interventions. Yet, studying the genetic origins of AN is challenging, in part due to its early age at onset and sexes imbalance, but perhaps also because people with AN often do not seek help for their condition (4,17), adding to the difficulty of obtaining sufficient sample sizes for genome-wide analyses. Still, two major genome-wide association studies (GWAS) on AN have been conducted by the Psychiatric Genomic Consortium (PGC)(19,20). The latest of these GWAS included 16,992 people with AN and 55,525 healthy controls, and identified eight genetic risk loci that reached genome-wide significance (19). Further, they showed that AN is genetically correlated with several psychiatric and metabolic traits, including body mass index, fat mass, and cholesterol levels. These findings have provided evidence that low weight may not only be a consequence of AN-related cognitions (i.e., fear of gaining weight, disturbed body image), but that it is possible that some metabolic traits predisposing to low weight may facilitate the development and maintenance of AN (19,20).

An aspect that is often overlooked in the study of the origins of AN is evolution. Indeed, the heritability estimate of 48-74% (21) suggest that some AN-associated genes must have been advantageous in recent human evolution, such as for example conferring advantages in terms of nutrition and reproduction (22–24). One may hypothesize that at some point in modern human history, it was beneficial to be able to stay physically active in order to travel and forage continuously in situations where food was scarce. Thus, some genes associated with AN may have conferred some sort of metabolic advantage in this sense such as higher tolerability to starvation, or different mechanisms of energy use allowing for increased physical activity (22–24). This would be somewhat consistent with what is seen in rodent models of AN, where animals subjected to time-restricted feeding and with access to a running wheel show increasing hyperactivity paired with a rapid reduction in food intake and body weight, eventually leading to death (23,25).

An evolutionary perspective on the genetics of AN may eventually also help better understanding the link between AN and puberty. It is possible that genes related to slower or faster sexual maturation may have been conserved to facilitate the human specie survival in less favourable environmental conditions (26,27). Since periods of growth such as childhood and adolescence are energy demanding, one may hypothesize that individuals who matured more quickly may have become less demanding more quickly in terms of resource use, saving resources to the benefit of their group. At the same time, since AN is often associated with amenorrhea, some authors have proposed that AN might have been a way to supress reproductive capacity when environmental conditions were not favourable, in order to increase reproductive success by saving energy for later, when more optimal conditions would be reached (22–24). This would be consistent with results from the few studies assessing the link between AN and sexual maturation, which suggest a possible delayed growth or sexual maturation in individuals with AN (28,29). It is also possible that genes impacting puberty in either direction may have been beneficial in terms of survival and reproduction depending on environmental scenarios, yet those genes might also confer an increased risk of EDs in today’s living condition.

One way to study such evolutionary adaptations is to use the genome of a very close relative of “modern humans” – Neanderthals. Archaeological evidence suggests that the two species co-existed on Earth for many years, but that at some point *Homo neanderthalensis* became extinct while *Homo sapiens* proliferated (30,31). Comparing the genomes of modern humans and Neanderthal might reveal *sapiens*-specific features that may have provided our specie with an evolutionary advantage (31). This has been studied in other psychiatric conditions such as schizophrenia (32) which has previously been linked to genes associated with creativity and language in modern humans (33,34), and there is evidence to suggest that the risk loci associated with schizophrenia may have been particularly subject to positive selection since the divergence of *Homo sapiens* and *Homo neanderthalensis* (32).

The overarching aim of the present study was to explore the genetic link between AN, BMI and age at menarche, and to apply an evolutionary perspective to gain a better understanding of the biology underlying AN. Specifically, our objectives were to (1) deepen knowledge of the genetic architecture of AN through its associations with BMI and age at menarche, and to identify novel gene variants shared between those traits, and (2) to gain a better understanding of the evolutionary origins of AN in general, and the identified variants in particular.

## Methods

### Samples

The present study uses publicly available summary statistics from GWAS of AN (19), BMI (35), and age at menarche (36). To avoid sample overlap, we selected GWAS that did not include participants from the UK Biobank for the BMI and age at menarche phenotypes. The AN GWAS was conducted by the Psychiatric Genomics Consortium and covered data from n = 16,992 individuals with AN and n = 55,525 healthy controls. The BMI GWAS was conducted by the GIANT consortium using data from n = 339,224 individuals. For age at menarche, we accessed a GWAS from the ReproGen Consortium performed in n = 182,416 individuals. All GWAS included individuals predominantly of European ancestry. Data collection for each GWAS was performed with participants’ written informed consent and with approval by the respective local Institutional Review Boards, as specified in the respective studies (19,35,36).

### Statistical Analyses

#### Genetic correlation

We calculated genetic correlations between AN and BMI, as well as between AN and age at menarche, using LD-score regression (v1.0.1), available at [https://github.com/bulik/ldsc](https://github.com/bulik/ldsc). Notably, these genetic correlations have been reported in previous studies (19,20), yet with different summary statistics, and therefore we recalculated the correlations to allow for direct comparison with our genetic overlap analysis.

#### Conjunctional false discovery rate

We used conjunctional FDR (conjFDR) analyses to investigate genetic overlap between AN and BMI and between AN and age at menarche. Detailed information about the conjFDR statistical framework can be found in previous publications (37,38), and the software can be accessed online at [https://github.com/precimed/pleiofdr/](https://github.com/precimed/pleiofdr/). We performed the analysis in MATLAB R2017b, and used the standard recommendation of FDR < 0.05 for our conjFDR threshold. Notably, the major histocompatibility complex (MHC) region was excluded before running conjFDR analyses (chr6:25,119,106–33,854,733).

#### Gene mapping and functional annotation

We uploaded the FDR statistics from our genetic overlap analysis to the Functional Mapping and Annotation (FUMA) platform (available here: [https://fuma.ctglab.nl/](https://fuma.ctglab.nl/)). This allowed us to functionally map and annotate the candidate SNPs we had identified, using positional, eQTL, and chromatin interaction mapping (39). FUMA aims to facilitate the identification of most likely causal variants from GWAS results (39). The FUMA SNP2GENE function annotates SNPs with functional categories, CADD scores (deleterious protein effect indicated by a CADD > 12.37 (40)), RegulomeDB scores (probable regulatory functions indicated by a score < 1F (41)) and chromatin states (more open chromatin state, indicated by a score between 1-7, suggest a region is more accessible for epigenetic regulatory mechanisms(42)). In the present study, locus numbers reflect their positional order from chromosome 1-22. Further, we used the GENE2FUNC function to identify enriched gene-sets and overrepresented biological pathways. In the present study, we considered SNPs with a conjFDR > 0.05 and a linkage disequilibrium (LD) R2 < 0.6 with each other as *independent significant SNPs*. Among these, SNPs with a R2 < 0.1 with each other were considered as *lead SNPs*. *Candidate SNPs* with a LD R2 ≥ 0.6 with the lead variants and with a conjFDR < 0.1 were submitted to FUMA. Lastly, novelty was assessed against related studies (19,20,43–46).

#### Neandertal selective sweep score

We used the Neandertal selective sweep (NSS) score to assess possible positive selection in humans at genomic loci associated with AN (R 4.2.2 ([www.r-project.org](http://www.r-project.org))). The NSS score can be downloaded from the UCSC genome browser ([http://genome.ucsc.edu](http://genome.ucsc.edu), ntSssZScorePMVar track (S-scores)), providing the likelihood index of positive selection in humans after the divergence of Neanderthals and humans (31,47). The NSS score measures the relative abundance of ancestral *vs.* non-ancestral alleles in Neandertals and humans, through the alignment of primates, Neanderthal and human consensus sequences – with a negative NSS score indicative of plausible positive selection in humans (31,47).

First, we used plink to clump candidate SNPs for the AN, BMI and age at menarche main GWAS ([https://github.com/precimed/python_convert](https://github.com/precimed/python_convert)). Then, we extracted NSS scores for all candidate SNPs identified in the AN, BMI and age at menarche GWAS, as well as for all candidate SNPs identified through our analysis of genetic overlap between them (conjFDR). This allowed us to examine the likelihood of positive selection (i.e., more negative NSS scores are associated with increased likelihood of positive selection) for the candidate SNPs associated with AN. Quantile-Quantile (QQ) plots were created to investigate the enrichment of more negative NSS scores in SNPs associated with AN. In addition, we graphically explored the distribution of NSS scores for candidate SNPs identified in previous AN, BMI and age at menarche GWAS, as well as candidate SNPs overlapping between the traits.

## Results

### Anorexia Nervosa genetically overlaps with Body Mass Index and Age at Menarche

We identified significant polygenic overlap between AN and BMI, and between AN and age at menarche. For both BMI and age at menarche, the proportion of non-null SNPs in AN increased with higher levels of these phenotypes, and *vice versa*. **Figure 1** illustrates the corresponding stratified Q-Q plots and shows an increasing enrichment as the statistical threshold becomes more stringent, supporting polygenic overlap between AN and BMI, and between AN and age at menarche.

![Figure 1:](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2023/08/15/2023.08.09.23293879/F1.medium.gif)

[Figure 1:](http://medrxiv.org/content/early/2023/08/15/2023.08.09.23293879/F1)

Figure 1: anorexia nervosa genetically overlaps with body mass index and age at menarche
Stratified quantile-quantile (QQ) plots displaying (A) increasing levels of SNP enrichment for AN conditioned on association *p*-values for BMI, and *vice versa*; (B) increasing levels of SNP enrichment for AN conditioned on association *p*-values for age at menarche, and *vice versa*

This overlap was further supported by statistically significant shared genomic loci between the phenotypes. Specifically, we identified 22 shared genomic risk loci between AN and BMI (67 independent significant SNPs), as well as nine shared genomic risk loci between AN and age at menarche (39 independent significant SNPs). Five loci overlapped between these two analyses (i.e., overlapped between AN and BMI and between AN and age at menarche), yielding 26 distinct loci overlapping with AN in total. To the best of our knowledge, 12 of these 26 distinct risk loci have not previously been reported to be associated with AN. **Figure 2** show the corresponding Manhattan plot for the overlapping loci, while **Supplementary Tables 1-2** provide further details on each of them. For comparison, genetic correlation analysis did not capture the overlap between AN and age at menarche (rg=-0.03; *p* = 0.35), and only identified a link between AN and BMI (rg=-0.21; *p* = 7.3×10-12). These correlation estimates are consistent with a previous report (19) and illustrate genetic overlap in the absence of genetic correlation (see (48) for review).

![Figure 2:](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2023/08/15/2023.08.09.23293879/F2.medium.gif)

[Figure 2:](http://medrxiv.org/content/early/2023/08/15/2023.08.09.23293879/F2)

Figure 2: Identification of loci shared between anorexia nervosa and body mass index, and anorexia nervosa and age at menarche
Manhattan plots presenting in color the significant loci identified as shared between (A) AN | BMI and; (B) AN | Menarche.

### Functional Annotation of Overlapping Loci

The majority of the SNPs identified in the genetic overlap analyses were located in intronic and intergenic regions (**Supplementary Figure 1**). For both analyses, variants were mapped to genes with enriched expression in brain tissues and in skeletal muscle (mostly downregulation; **Supplementary** Figure 2-3). Functions of these genes included neurodevelopmental, metabolic and cellular processes (**Supplementary Figures 4-5**). In both analyses, we identified genes linked to sleep and chronotype, cognition and mental health, anthropometric traits, reproduction, regular attendance at a gym or sport club and cardiovascular health (**Supplementary Figure 6-7**).

We identified *ARNTL* (more commonly known as *BMAL1*), known for its importance in the regulation of sleep, weight and eating behavior, as part of a locus shared between AN and age at menarche (locus 6) and between AN and BMI (locus 16). Another new locus shared between AN and age at menarche (locus 5) and between AN and BMI (locus 12) was mapped to *PARD3*, which is involved in various neurodevelopmental, cellular, metabolic and developmental processes and has been linked to schizophrenia.

In addition, the analysis between AN and BMI identified seven new loci for AN that did not overlap with those shared between AN and age at menarche. Most of the genes from the 22 overlapping loci between AN and BMI were related to neurodevelopmental processes (*NEGR1*, *BCL11A, SLC25A12, KIAA1593, PCDH17, ADARB1*) and overlapped with mental health-related phenotypes, such as mood swings, schizophrenia, neuroticism and risk-taking – which is supported by previous studies on their expression patterns (e.g., *BCL11A* (49), *PCDH17* (50,51)). Further mental health problems previously identified as linked to the genes overlapping between AN and BMI include bulimia nervosa and major depressive disorders (*NEGR1*, (52,53)) autism spectrum disorders (*SLC25A12,* (54)) and suicide attempts (*ADARB1* (55)).

Finally, the analysis between AN and age at menarche identified two new loci for AN that did not overlap with those identified in the analysis between AN and BMI. Of these, locus 4 was close to the *NR5A1* gene, which is involved in gonad development, hormonal mediated signalling pathways and aging. Further, genes shared between AN and age at menarche were largely related to autoimmune/autoinflammatory disorders. Gut microbiota also came out as linked to genes shared between AN and age at menarche.

### Possible Positive Selection of Loci Associated With Anorexia Nervosa

Our analysis of Neanderthal Selective Sweep scores revealed that gene variants related to AN may have been positively selected in recent human evolution. The QQ-plot in **Figure 3** illustrates enrichment of AN-related SNPs for more negative NSS scores, where negative NSS scores point toward recent positive evolutionary selection.

![Figure 3:](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2023/08/15/2023.08.09.23293879/F3.medium.gif)

[Figure 3:](http://medrxiv.org/content/early/2023/08/15/2023.08.09.23293879/F3)

Figure 3: Comparison to Neanderthal genome suggests recent evolutionary selection for SNPs associated with anorexia nervosa
Quantile-Quantile (QQ) plot showing the enrichment of more negative Neanderthal Selective Sweep scores in SNPs associated with AN, pointing toward recent positive evolutionary selection.

It should be noted that after merging the AN GWAS summary statistic with NSS score data we retained mostly SNPs with negative NSS scores, warranting caution to the deployed NSS cut-off at 0. We therefore assessed NSS patterns for BMI and age at menarche, as well as for specific genes overlapping between these phenotypes and AN. First, we compared the distribution of NSS scores for SNPs associated with AN, BMI and those identified as shared between AN and BMI (**Figure 4A**). We found that AN candidate SNPs were enriched for negative NSS scores (mean: -4.78, sd: 1.35) compared to BMI candidate SNPs (mean: -3.43, sd: 1.9), with candidate SNPs overlapping between AN and BMI standing in-between (mean: -3.79, sd: 1.73). We repeated this approach for AN and age at menarche and found similar patterns as above, that is more negative NSS scores for AN candidate SNPs compared to SNPs associated with age at menarche (mean: -3.93, sd: 1.62), and with SNPs overlapping between AN and age at menarche (mean: -4.47, sd: 1.51) standing in-between (**Figure 4B**). To put these results into perspective, we performed a conjFDR analysis to identify genetic overlap between BMI and age at menarche, merged our results with the NSS score data and compared the NSS score distributions for candidate SNPs associated with BMI, age at menarche or BMI and age at menarche. We found that the distribution of NSS scores between BMI and age at menarche was similar and less skewed to the left (i.e. less negative NSS values) than what we observed for candidate SNPs associated with AN (**Figure 4C**), further supporting that recent positive selection played a particular role in AN associated gene variants.

![Figure 4:](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2023/08/15/2023.08.09.23293879/F4.medium.gif)

[Figure 4:](http://medrxiv.org/content/early/2023/08/15/2023.08.09.23293879/F4)

Figure 4: Compared to body mass index and age at menarche, SNPs associated with anorexia nervosa appear to be under stronger evolutionary selection
Distribution of Neanderthal Selective Sweep scores for candidate SNPs associated with (A) AN, BMI and those shared between AN | BMI; (B) AN, age at menarche and those shared between AN | Menarche; (C) BMI, age at menarche, and those shared between BMI | Menarche. Negative Neanderthal Selective Sweep scores point toward recent positive evolutionary selection.

### Evolutionary Placement of AN-associated Genes

Of all the identified loci, we investigated whether some were more likely to be subject to recent evolutionary dynamics. NSS score values were available for 113 of the 275 candidate SNPs identified as overlapping between AN and BMI (41%), and for 38 of the 92 candidate SNPs overlapping between AN and age at menarche (41%). The average NSS score of these candidate SNPs was calculated for each locus, and is reported in **Supplementary Table 3**. We next calculated mean NSS scores per gene, allowing us to rank genes by their evolutionary signal, as shown in **Figure 5**.

![Figure 5:](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2023/08/15/2023.08.09.23293879/F5.medium.gif)

[Figure 5:](http://medrxiv.org/content/early/2023/08/15/2023.08.09.23293879/F5)

Figure 5: Genes associated with anorexia nervosa, highlighted from an evolutionary perspective
Genes mapped to the loci associated with AN (top panel), AN | BMI (middle panel) or AN | Menarche (bottom panel). The x-axis represents Neanderthal Selective Sweep scores, with more negative Neanderthal Selective Sweep scores pointing toward recent positive evolutionary selection. Neanderthal Selective Sweep score per gene was determined as an average of the SNPs from which the gene was mapped. Location on y-axis is random (jitter) and was used for legibility of otherwise overlapping text. Darker colours are used to highlight genes with the lowest mean NSS scores.

The gene with the most negative NSS score was *ACA59*, which mapped to a locus identified in the AN GWAS (chromosome 11). The gene encodes a small nucleolar RNA (snoRNA), SNORA59A. We did not find any association with biological functions or GWAS phenotypes via FUMA for this specific gene, nor did we find relevant literature on its function. In the analysis of genetic overlap between AN and BMI, the gene with most negative NSS score was *GNB1* (chromosome 1), which is involved in cell-cell signaling. Mutations in *GNB1* have been linked to growth and (neuro)developmental delay, as well as gastrointestinal problems (56). In the overlap between AN and age at menarche, *USP4* (chromosome 3) had the most negative NSS score.

One gene with strong negative NSS found in all three analyses (AN GWAS, AN|BMI overlap, AN|Menarche overlap) was *RBM6*, which has previously been linked to body fat distribution and age at menarche, and is also known for its role in sleep and chronotype. Several other genes in the same locus (chromosome 3) also showed strongly negative NSS but were not necessarily mapped in all three analyses. These were *CCDC36* (AN, AN|Menarche), *DAG1* (AN, AN|Menarche), *NCKIPSD* (AN), *USP4* (AN, AN|Menarche) and *USP19* (AN). *USP4* and *DAG1* are known for their roles in diverse processes such as sleep and chronotype, autoimmune/autoinflammatory disorders, cognition and metabolic function. *DAG1* also has roles in heart biology and gut microbiota, and has been implicated in functions relevant to neurodevelopment, such as synapse organization, neuronal differentiation and guidance, neurogenesis, glial cell differentiation and myelination.

A common pattern that emerged among many genes with strongly negative NSS was the implication of chronotype as an associated process. For example, *GRP7-ASB3* has been linked to chronotype and excessive daytime sleepiness, while *CAMKV*, *IP6K2* and *RHOA* have been associated with morning-ness, and *MONIA, RHOA, TCTA, AMT, NICN1, CAMKV* relate to sleep duration.

## Discussion

The present study established polygenic overlap between AN, BMI and age at menarche and through its overlap identified 12 novel loci associated with AN. Further, we found that the genetic risk loci associated with AN may have been subject to recent evolutionary selection. The current findings deepen our knowledge of the genetics of AN and may help to better understand the biological mechanisms involved in this debilitating disorder.

Our analysis of genetic overlap between AN and BMI identified 22 significantly overlapping risk loci. Previous research has already established the inverse genetic correlation between AN and BMI, and that fat mass shows one of the strongest genetic correlations with AN (19). These correlations and the here identified overlap support that energy balance may be impaired in people with AN. At the same time, the association between AN and BMI is not solely linked to energy balance, as a large proportion of the genes identified as overlapping between AN and BMI are relevant to brain and neurodevelopmental functions. Among these genes, the identification of *NEGR1* as overlapping between AN and BMI (locus 2) is noteworthy, as *NEGR1* is involved in neurodevelopment and has previously been associated with obesity and age at menarche (35,57), as well as other psychiatric disorders (52,53). In an epigenome-wide study, *NEGR1* was found to be differentially methylated in people with active AN compared to people in remission from AN or healthy controls (58), further supporting its possible role in AN. Overall, our findings reinforce that studying EDs and overweight/obesity together will likely be beneficial, as the underlying (neuro)biology of these conditions may overlap (59).

Puberty is thought to be an important time period for the development of EDs and may underline some of the sex differences in the prevalence and clinical presentation of EDs (12). Indeed, we here identified nine significant loci overlapping between AN and age at menarche. This overlap is noteworthy given their weak, non-significant genetic correlation, as also previously reported (19). These results may indicate that some of the variants may have an opposite effect on AN and age at menarche. For example, some genes associated with an increased risk of AN may be associated with an earlier age at menarche, while others may be associated with a later age at menarche. In the future, it will be insightful to assess whether the association between AN and age at menarche varies by AN subtypes (e.g. restrictive *vs.* binge/purge or early *vs.* late onset). In this regard, a previous study found that an earlier age at menarche was genetically correlated to an earlier age at onset of AN in people with early-onset AN (45). However, in this previous study, age at menarche was not associated with the risk of AN or with late-onset AN (45). These findings may be somewhat surprising given previous evidence of delayed growth and sexual maturation in youths with AN (28,29). Yet, this may also point to the existence of different AN subtypes with potentially diverse biological underpinnings, especially given that the literature on AN and age at menarche remains relatively scarce, with few authors considering AN subtypes, and sample sizes being often limited. Further research is therefore warranted, ideally including puberty-related variables as a primary focus (and not just as confounders).

A new finding from the present study is that several genes overlapping between AN, BMI and age at menarche have in common that they are related to circadian-related phenotypes. These include many of the genes that were most likely to have been under recent evolutionary selection. Relatedly, the interaction between circadian rhythms, sleep and food intake has been established, but not well studied in the context of EDs. Existing evidence suggests that sleep problems are common in people with EDs (60–63) and that sleep disruptions can worsen the severity of AN (64). Poor sleep and altered circadian rhythms have also been linked to metabolic dysregulation and obesity (65,66), as well as to woman’s reproductive health (67,68). One might therefore wonder whether the circadian cycle might be a key element linking AN, BMI and puberty, and whether this phenotype might have conferred some sort of evolutionary advantage. Depending on environmental conditions, being more or less “restless” might have impacted the risk to suffer from malnutrition and might have affected the chances of reproductive success. More studies are needed to answer this question, but it appears that sleep problems and circadian cycles may be promising avenues of research in people with AN, or even more broadly in a spectrum of EDs, but also in the context of women’s reproductive health. A better understanding of the interaction between these phenotypes may provide an interesting starting point for the development of future interventions to prevent and treat EDs and related problems.

Adding to the importance of the circadian cycle, one of the new loci we have identified overlapped between AN, BMI, and age at menarche and mapped to the *BMAL1* gene (synonym: *ARNTL*), a major component of the circadian clock system (69,70). In animal models, knock-out or reduced expression of *BMAL1* have been associated with altered circadian rhythms, body weight, alterations in locomotor activity and eating behaviours, delayed puberty and infertility, cognitive deficits and shortened lifespan (67,68,70). To the best of our knowledge, this is the first report of an association between *BMAL1* and AN; however, the gene has previously been associated with both BMI and age at menarche (57,71). *BMAL1* has been linked to several factors previously implicated in EDs and obesity (19,20,58,72–75), including various facets of the appetite regulation system and energy metabolism, such as adipogenesis, ghrelin, insulin, leptin and adiponectin pathways (76–78). Further, there may be a link between *BMAL1*, appetite pathways and reproduction, as *BMAL1* silencing has been linked to reduced estrogen synthesis in granulosa cells via reduced leptin secretion (79).

Other mechanisms linked to the circadian cycle may be relevant to AN. For example, one study found that sleep deprivation in mice was associated with both increased food intake and decreased body weight (80). The latter study also found an association of sleep deprivation with changes in the gut microbiota, systemic inflammation and lower *BMAL1* levels in the colon (80). There are possibly parallels to be drawn with AN, as there is a bidirectional relationship between many autoimmune and autoinflammatory disorders affecting the gastrointestinal system and EDs (81,82), and alterations in the microbiota-gut-brain axis have been linked to AN (83). The present study may provide additional evidence for a link between these traits, as we identified genes related to the gut microbiota as overlapping between AN and age at menarche (e.g., *RHOA, AMT, DAG1, BSN*) and genes related to gastrointestinal autoimmune/autoinflammatory disorders (e.g., *RHOA, AMT, DAG1, BSN, MST1, RNF123*) as overlapping between AN, BMI and age at menarche.

Studying AN from an evolutionary perspective revealed that some of the risk loci for AN might have had a favourable effect on the survival of modern humans, as suggested by an enrichment in negative NSS scores. Although more studies are needed to confirm this finding, this result would align with studies in other fields that have suggested various metabolic adaptations after exposure to developmental adversity. For example, longitudinal studies on the Dutch Famine and Chinese Hunger cohorts have shown that individuals who were exposed to hunger during pregnancy had a higher risk of long-term health problems, including psychiatric and metabolic disorders (84–86). Some have hypothesised that these effects could be mediated by epigenetic mechanisms, which allow for interactions between genes and environment (87). It could be further hypothesised that similar adaptations may have occurred *genetically*, over many generations, to favour *homo sapiens* survival. Along the way, some adaptations that conferred a metabolic or reproductive advantage may now also increase the risk of certain disorders, including AN. Still, we cannot exclude that the enrichment of negative NSS scores at SNPs related to AN may have other explanations, for example that some of the SNPs associated with AN did not confer an evolutionary advantage but may have been inherited along with SNPs with beneficial functions, through linkage disequilibrium. Nevertheless, our comparison of the distribution of NSS scores between AN, BMI and menarche SNPs suggests particularly pronounced effects for AN from an evolutionary point of view. Some of the biology underlying these evolutionary patterns may still need to be explored. For example, we here found *ACA59*, encoding a small non-coding RNA (snoRNA), as the gene with strongest signal in terms of recent evolutionary selection, yet little is known about the function of this particular gene in human physiology. Still, snoRNAs are known to be evolutionarily conserved across various organisms (88), and evidence from rodent models suggests that the expression patterns of some snoRNAs may be under circadian regulation (89,90). This is interesting given our findings that genes shared between AN, BMI and age at menarche are related to the circadian cycle, and warrants further investigations.

The present study has some important strengths, including the use of statistical methods that allowed us to explore genetic pleiotropy and the use of Neanderthal genomic data to employ an evolutionary lens on AN. Yet, some limitations need to be considered. Merging the GWAS data with the NSS score data left us with several SNPs or even loci for which NSS scores were not available and thus some AN-associated signal could not be explored in terms of evolution. Furthermore, most retained SNPs had a negative NSS score (pointing toward recent selection). We circumvented this potential issue by comparing NSS distributions between phenotypes, all of which had overall more negative than positive NSS scores. Moreover, the current data did not allow us to study subtypes of AN, such as restrictive or binge-purge subtypes, nor between men and women, although it is possible that there are relevant differences between these groups. Indeed, our results of genetic overlap between AN and age at menarche despite a correlation close to zero indicate diverging effect directions that cancel each other out in a correlation framework. One may speculate that different effect directions in different AN subtypes could be one possible cause, such as genes relevant to one subtype associated with earlier menarche and genes relevant to another subtype linked with later menarche. Finally, an important perspective for future work is the need for longitudinal studies that will allow to explore the developmental course underlying the association between AN, BMI, and age at menarche.

In conclusion, the present study has identified genes that overlap between AN, BMI and age at menarche, strengthening the association between these phenotypes, and found that associated genes may have been undergoing recent evolutionary selection. We identified genes relevant to circadian rhythms in the association signal between AN, BMI and age at menarche, and with Neanderthal scores pointing at recent selection. Our study emphasizes the utility of employing an evolutionary lens in psychiatric genetics and the results identify candidates for future studies, such as detailed investigations into the genetic link between chronobiology and AN. As such, analysis of pleiotropy and evolutionary context may go hand in hand in further deciphering the mechanisms underlying this debilitating disorder.

## Supporting information

Supplementary Figures 1-7 [[supplements/293879_file02.pdf]](pending:yes)

Supplementary Tables 1-3 [[supplements/293879_file03.xlsx]](pending:yes)

## Data Availability

All data are available online from the respective consortia

[https://www.reprogen.org/data\_download.html](https://www.reprogen.org/data_download.html) 

[https://portals.broadinstitute.org/collaboration/giant/index.php/GIANT\_consortium\_data\_files](https://portals.broadinstitute.org/collaboration/giant/index.php/GIANT_consortium_data_files) 

[https://pgc.unc.edu/for-researchers/download-results/](https://pgc.unc.edu/for-researchers/download-results/) 

## Funding

This work was funded by the Research Council of Norway (grant #323961 to TK). TK is a member of the Machine Learning Cluster of Excellence, EXC number 2064/1 – Project number 39072764. EB is the recipient of postdoctoral training awards from the Canadian Institutes of Health Research (#489915) and from the Fonds de Recherche du Québec – Santé (#329028).

## Conflicts of Interest

The authors have no conflict of interest to declare.

## Acknowledgements

This work was undertaken on the Tjeneste for Sensitive Data (TSD) facilities which is operated/developed by University of Oslo TSD service group (IT-Department), and owned by the University of Oslo as well as on resources provided by the National Infrastructure for High Performance Computing and Data Storage in Norway (UNINETT Sigma2). We would like to thank the three consortia (PGC, GIANT and ReproGen) from which we obtained summary statistics, and their respective research participants, for making this work possible.

*   Received August 9, 2023.
*   Revision received August 9, 2023.
*   Accepted August 15, 2023.


*   © 2023, Posted by Cold Spring Harbor Laboratory

This pre-print is available under a Creative Commons License (Attribution-NoDerivs 4.0 International), CC BY-ND 4.0, as described at [http://creativecommons.org/licenses/by-nd/4.0/](http://creativecommons.org/licenses/by-nd/4.0/)

## References

1.  1.Eddy KT, Tabri N, Thomas JJ, Murray HB, Keshaviah A, Hastings E, et al. Recovery from anorexia nervosa and bulimia nervosa at 22-year follow-up. J Clin Psychiatry. 2017;78(2):184–9.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.4088/JCP.15m10393&link_type=DOI) 

2.  2.Treasure J, Duarte TA, Schmidt U. Eating disorders. The Lancet. 2020;395(10227):899–911.
    
    

3.  3.Schaumberg K, Welch E, Breithaupt L, Hübel C, Baker JH, Munn-Chernoff MA, et al. The science behind the academy for eating disorders’ nine truths about eating disorders. European Eating Disorders Review. 2017;25(6):432–50.
    
    

4.  4.Stone KD, Dimitropoulos G, MacMaster FP. Food for thought: A dissonance between healthcare utilization costs and research funding for eating disorders in Canada. J Can Acad Child Adolesc Psychiatry. 2021;30(3):197–203.
    
    

5.  5.Arcelus J, Mitchell AJ, Wales J, Nielsen S. Mortality rates in patients with anorexia nervosa and other eating disorders. A meta-analysis of 36 studies. Arch Gen Psychiatry. 2011;68(7):724–31.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1001/archgenpsychiatry.2011.74&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21727255&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000292356600008&link_type=ISI) 

6.  6.Swanson SA, Crow SJ, Le Grange D, Swendsen J, Merikangas KR. Prevalence and correlates of eating disorders in adolescents. Arch Gen Psychiatry. 2011;68(7):714– 23.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1001/archgenpsychiatry.2011.22&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21383252&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000292356600007&link_type=ISI) 

7.  7.Lenroot RK, Giedd JN. Sex differences in the adolescent brain. Brain Cogn. 2010;72(1):46.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.bandc.2009.10.008&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=19913969&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000274128400006&link_type=ISI) 

8.  8.Berkowitz SA, Witt AA, Gillberg C, Råstam M, Wentz E, Lowe MR. Childhood body mass index in adolescent-onset anorexia nervosa. International Journal of Eating Disorders. 2016;49(11):1002–9.
    
    

9.  9.Marion M, Lacroix S, Caquard M, Dreno L, Scherdel P, Guen CGL, et al. Earlier diagnosis in anorexia nervosa: better watch growth charts! Journal of Eating Disorders. 2020;8:42.
    
    

10. 10.Stice E, Desjardins CD, Rohde P. Young women who develop anorexia nervosa exhibit a persistently low premorbid body weight on average: A longitudinal investigation of an important etiologic clue. J Psychopathol Clin Sci. 2022;131(5):479– 92.
    
    

11. 11.Lebow J, Sim LA, Kransdorf LN. Prevalence of a history of overweight and obesity in adolescents with restrictive eating disorders. Journal of Adolescent Health. 2015;56(1):19–24.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.jadohealth.2014.06.005&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=25049202&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

12. 12.Breton É, Juster RP, Booij L. Gender and sex in eating disorders: A narrative review of the current state of knowledge, research gaps, and recommendations. Brain and Behavior. 2023;13(4):e2871.
    
    

13. 13.Aparicio-Martinez P, Perea-Moreno AJ, Martinez-Jimenez MP, Redel-Macías MD, Pagliari C, Vaquero-Abellan M. Social media, thin-ideal, body dissatisfaction and disordered eating attitudes: An exploratory analysis. Int J Environ Res Public Health. 2019;16(21): 4177
    
    

14. 14.Jenkins ZM, Chait LM, Cistullo L, Castle DJ. A comparison of eating disorder symptomatology, psychological distress and psychosocial function between early, typical and later onset anorexia nervosa. J Eat Disord. 2020;8(1):56.
    
    

15. 15.Steinhausen HC, Villumsen MD, Hørder K, Winkler LAD, Bilenberg N, Støving RK. Comorbid mental disorders during long-term course in a nationwide cohort of patients with anorexia nervosa. Int J Eat Disord. 2021;54(9):1608–18.
    
    

16. 16.Dobrescu SR, Dinkler L, Gillberg C, Råstam M, Gillberg C, Wentz E. Anorexia nervosa: 30-year outcome. Br J Psychiatry. 2020;216(2):97–104.
    
    

17. 17.Silén Y, Sipilä PN, Raevuori A, Mustelin L, Marttunen M, Kaprio J, et al. Detection, treatment, and course of eating disorders in Finland: A population-based study of adolescent and young adult females and males. Eur Eat Disord Rev. 2021;29(5):720–732
    
    

18. 18.Castro TF, Miller K, Araújo MX, Brandão I, Torres S. Emotional processing in recovered anorexia nervosa patients: A 15 year longitudinal study. Eur Eat Disord Rev. 2021;29(6):955–68.
    
    

19. 19.Watson HJ, Yilmaz Z, Thornton LM, Hübel C, Coleman JRI, Gaspar HA, et al. Genome-wide association study identifies eight risk loci and implicates metabo-psychiatric origins for anorexia nervosa. Nat Genet. 2019;51(8):1207–14.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/S41588-019-0439-2&link_type=DOI) 

20. 20.Duncan L, Yilmaz Z, Gaspar H, Walters R, Goldstein J, Anttila V, et al. Significant locus and metabolic genetic correlations revealed in genome-wide association study of anorexia nervosa. Am J Psychiatry. 2017;174(9):850–8.
    
    

21. 21.Yilmaz Z, Hardaway JA, Bulik CM. Genetics and epigenetics of eating disorders. Adv Genomics Genet. 2015;5:131–50.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.2147/AGG.S55776&link_type=DOI) 

22. 22.Gatward N. Anorexia nervosa: an evolutionary puzzle. Eur Eat Disord Rev. 2007;15(1):1–12.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1002/erv.718&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=17676667&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000244492200001&link_type=ISI) 

23. 23.Guisinger S. Adapted to flee famine: adding an evolutionary perspective on anorexia nervosa. Psychol Rev. 2003;110(4):745–61.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1037/0033-295X.110.4.745&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=14599241&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

24. 24.Mayhew AJ, Pigeyre M, Couturier J, Meyre D. An evolutionary genetic perspective of eating disorders. Neuroendocrinology. 2017;106(3):292–306.
    
    

25. 25.Zhang J, Dulawa SC. The utility of animal models for studying the metabo-psychiatric origins of anorexia nervosa. Frontiers in Psychiatry. 2021;12:711181.
    
    

26. 26.Papadimitriou A. The evolution of the age at menarche from prehistorical to modern times. Journal of Pediatric and Adolescent Gynecology. 2016;29(6):527–30.
    
    

27. 27.Gillette MT, Folinsbee KE. Early menarche as an alternative reproductive tactic in human females: An evolutionary approach to reproductive health issues. Evol Psychol. 2012;10(5):147470491201000500.
    
    

28. 28.Neale J, Pais SMA, Nicholls D, Chapman S, Hudson LD. What are the effects of restrictive eating disorders on growth and puberty and are effects permanent? A systematic review and meta-analysis. J Adolesc Health. 2020;66(2):144–56.
    
    

29. 29.Clarke J, Peyre H, Alison M, Bargiacchi A, Stordeur C, Boizeau P, et al. Abnormal bone mineral density and content in girls with early-onset anorexia nervosa. J Eat Disord. 2021;9:9.
    
    

30. 30.Hublin JJ, Spoor F, Braun M, Zonneveld F, Condemi S. A late Neanderthal associated with Upper Palaeolithic artefacts. Nature. 1996;381(6579):224–6.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/381224a0&link_type=DOI) 
    
    [GeoRef](http://medrxiv.org/lookup/external-ref?access_num=1996042394&link_type=GEOREF) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=8622762&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=A1996UL24900050&link_type=ISI) 

31. 31.Green RE, Krause J, Briggs AW, Maricic T, Stenzel U, Kircher M, et al. A draft sequence of the Neandertal genome. Science. 2010;328(5979):710–22.
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6Mzoic2NpIjtzOjU6InJlc2lkIjtzOjEyOiIzMjgvNTk3OS83MTAiO3M6NDoiYXRvbSI7czo1MDoiL21lZHJ4aXYvZWFybHkvMjAyMy8wOC8xNS8yMDIzLjA4LjA5LjIzMjkzODc5LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

32. 32.Srinivasan S, Bettella F, Mattingsdal M, Wang Y, Witoelar A, Schork AJ, et al. Genetic markers of human evolution are enriched in schizophrenia. Biol Psychiatry. 2016;80(4):284–92.
    
    

33. 33.Crow TJ. Is schizophrenia the price that Homo sapiens pays for language? Schizophr Res. 1997;28(2–3):127–41.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/S0920-9964(97)00110-2&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=9468348&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000071234900005&link_type=ISI) 

34. 34.Wynn T, Coolidge FL. The implications of the working memory model for the evolution of modern cognition. Int J Evol Biol. 2011;2011:741357.
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21716664&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

35. 35.Locke AE, Kahali B, Berndt SI, Justice AE, Pers TH, Day FR, et al. Genetic studies of body mass index yield new insights for obesity biology. Nature. 2015;518(7538):197–206.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/nature14177&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=25673413&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

36. 36.Perry JRB, Day F, Elks CE, Sulem P, Thompson DJ, Ferreira T, et al. Parent-of-origin-specific allelic associations among 106 genomic loci for age at menarche. Nature. 2014;514(7520):92–7.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/nature13545&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=25231870&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000342420800042&link_type=ISI) 

37. 37.Andreassen OA, Thompson WK, Schork AJ, Ripke S, Mattingsdal M, Kelsoe JR, et al. Improved detection of common variants associated with schizophrenia and bipolar disorder using pleiotropy-informed conditional false discovery rate. PLoS Genet. 2013;9(4):e1003455.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pgen.1003455&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=23637625&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

38. 38.Smeland OB, Frei O, Shadrin A, O’Connell K, Fan CC, Bahrami S, et al. Discovery of shared genomic loci using the conditional false discovery rate approach. Hum Genet. 2020;139(1):85–94.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s00439-019-02060-2&link_type=DOI) 

39. 39.Watanabe K, Taskesen E, van Bochoven A, Posthuma D. Functional mapping and annotation of genetic associations with FUMA. Nat Commun. 2017;8(1):1826.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41467-017-01261-5&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

40. 40.Rentzsch P, Witten D, Cooper GM, Shendure J, Kircher M. CADD: predicting the deleteriousness of variants throughout the human genome. Nucleic Acids Research. 2019;47(D1):D886–94.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/nar/gky1016&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

41. 41.Boyle AP, Hong EL, Hariharan M, Cheng Y, Schaub MA, Kasowski M, et al. Annotation of functional variation in personal genomes using RegulomeDB. Genome Res. 2012;22(9):1790–7.
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NjoiZ2Vub21lIjtzOjU6InJlc2lkIjtzOjk6IjIyLzkvMTc5MCI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDIzLzA4LzE1LzIwMjMuMDguMDkuMjMyOTM4NzkuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

42. 42.Roadmap Epigenomics Consortium, Kundaje A, Meuleman W, Ernst J, Bilenky M, Yen A, et al. Integrative analysis of 111 reference human epigenomes. Nature. 2015;518(7539):317–30.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/nature14248&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=25693563&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

43. 43.Peyrot WJ, Price AL. Identifying loci with different allele frequencies among cases of eight psychiatric disorders using CC-GWAS. Nat Genet. 2021;53(4):445–54.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41588-021-00787-1&link_type=DOI) 

44. 44.Genomic relationships, novel loci, and pleiotropic mechanisms across eight psychiatric disorders. Cell. 2019;179(7):1469–1482.e11.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.cell.2019.11.020&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=31835028&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

45. 45.Watson HJ, Thornton LM, Yilmaz Z, Baker JH, Coleman JRI, Adan RAH, et al. Common genetic variation and age of onset of anorexia nervosa. Biol Psychiatry Glob Open Sci. 2021;2(4):368–78.
    
    

46. 46.Hinney A, Kesselmeier M, Jall S, Volckmar AL, Föcker M, Antel J, et al. Evidence for three genetic loci involved in both anorexia nervosa risk and variation of body mass index. Mol Psychiatry. 2017;22(2):192–201.
    
    

47. 47.Burbano HA, Hodges E, Green RE, Briggs AW, Krause J, Meyer M, et al. Timing of human protein evolution as revealed by massively parallel capture of Neandertal nuclear DNA sequences. Science. 2010;328(5979):723–5.
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6Mzoic2NpIjtzOjU6InJlc2lkIjtzOjEyOiIzMjgvNTk3OS83MjMiO3M6NDoiYXRvbSI7czo1MDoiL21lZHJ4aXYvZWFybHkvMjAyMy8wOC8xNS8yMDIzLjA4LjA5LjIzMjkzODc5LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

48. 48.Smeland OB, Frei O, Dale AM, Andreassen OA. The polygenic architecture of schizophrenia — rethinking pathogenesis and nosology. Nat Rev Neurol. 2020;16(7):366–79.
    
    

49. 49.Chen J, Cao H, Kaufmann T, Westlye LT, Tost H, Meyer-Lindenberg A, et al. Identification of reproducible BCL11A alterations in schizophrenia through individual-level prediction of coexpression. Schizophr Bull. 2020;46(5):1165–71.
    
    

50. 50.Chang H, Hoshina N, Zhang C, Ma Y, Cao H, Wang Y, et al. The protocadherin 17 gene affects cognition, personality, amygdala structure and function, synapse development and risk of major mood disorders. Mol Psychiatry. 2018;23(2):400–12.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/mp.2016.231&link_type=DOI) 

51. 51.Dean B, Keriakous D, Scarr E, Thomas EA. Gene expression profiling in Brodmann’s area 46 from subjects with schizophrenia. Aust N Z J Psychiatry. 2007;41(4):308–20.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1080/00048670701213245&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=17464717&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000245948700002&link_type=ISI) 

52. 52.Gamero-Villarroel C, González LM, Gordillo I, Carrillo JA, García-Herráiz A, Flores I, et al. Impact of NEGR1 genetic variability on psychological traits of patients with eating disorders. Pharmacogenomics J. 2015;15(3):278–83.
    
    

53. 53.Wang X, Cheng W, Zhu J, Yin H, Chang S, Yue W, et al. Integrating genome-wide association study and expression quantitative trait loci data identifies NEGR1 as a causal risk gene of major depression disorder. Journal of Affective Disorders. 2020;265:679–86.
    
    

54. 54.Qiu S, Qiu Y, Li Y, Cong X. Genetics of autism spectrum disorder: an umbrella review of systematic reviews and meta-analyses. Transl Psychiatry. 2022;12:249.
    
    

55. 55.Karanović J, Ivković M, Jovanović VM, Šviković S, Pantović-Stefanović M, Brkušanin M, et al. Effect of childhood general traumas on suicide attempt depends on TPH2 and ADARB1 variants in psychiatric patients. J Neural Transm (Vienna*).* 2017;124(5):621–9.
    
    

56. 56.Hemati P, Revah-Politi A, Bassan H, Petrovski S, Bilancia CG, Ramsey K, et al. Refining the phenotype associated with GNB1 mutations: Clinical data on 18 newly identified patients and review of the literature. American Journal of Medical Genetics. 2018;176(11):2259–75.
    
    

57. 57.Elks CE, Perry JRB, Sulem P, Chasman DI, Franceschini N, He C, et al. Thirty new loci for age at menarche identified by a meta-analysis of genome-wide association studies. Nat Genet. 2010;42(12):1077–85.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/ng.714&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21102462&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

58. 58.Steiger H, Booij L, Thaler L, St-Hilaire A, Israël M, Casey KF, et al. DNA methylation in people with anorexia nervosa: Epigenome-wide patterns in actively ill, long-term remitted, and healthy-eater women. The World Journal of Biological Psychiatry. 2022;24(3): 254–259.
    
    

59. 59.Breton E, Fotso Soh J, Booij L. Immunoinflammatory processes: Overlapping mechanisms between obesity and eating disorders? Neuroscience & Biobehavioral Reviews. 2022;138:104688.
    
    

60. 60.Baglioni C, Nanovska S, Regen W, Spiegelhalder K, Feige B, Nissen C, et al. Sleep and mental disorders: A meta-analysis of polysomnographic research. Psychol Bull. 2016;142(9):969–90.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1037/bul0000053&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

61. 61.Asaad Abdou T, Esawy HI, Abdel Razek Mohamed G, Hussein Ahmed H, Elhabiby MM, Khalil SA, et al. Sleep profile in anorexia and bulimia nervosa female patients. Sleep Med. 2018;48:113–6.
    
    

62. 62.Menculini G, Brufani F, Del Bello V, Moretti P, Tortorella A. Circadian rhythms disruptions and eating disorders: clinical impact and possible psychopathological correlates. Psychiatr Danub. 2019;31(Suppl 3):497–502.
    
    

63. 63.Allison KC, Spaeth A, Hopkins CM. Sleep and eating disorders. Curr Psychiatry Rep. 2016;18(10):92.
    
    

64. 64.Malcolm A, Toh WL, Crocker K, Phillipou A. The contribution of sleep to anorexia nervosa severity. Eat Weight Disord. 2022;27(4):1563–8.
    
    

65. 65.Depner CM, Stothard ER, Wright KP. Metabolic consequences of sleep and circadian disorders. Curr Diab Rep. 2014;14(7):507.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s11892-014-0507-z&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=24816752&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

66. 66.Spiegel K, Tasali E, Leproult R, Van Cauter E. Effects of poor and short sleep on glucose metabolism and obesity risk. Nat Rev Endocrinol. 2009;5(5):253–61.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/nrendo.2009.23&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=19444258&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000265854100008&link_type=ISI) 

67. 67.Jiang Y, Li S, Xu W, Ying J, Qu Y, Jiang X, et al. Critical roles of the circadian transcription factor BMAL1 in reproductive endocrinology and fertility. Front Endocrinol (Lausanne*).* 2022;13:818272.
    
    

68. 68.Ono M, Ando H, Daikoku T, Fujiwara T, Mieda M, Mizumoto Y, et al. The circadian clock, nutritional signals and reproduction: A close relationship. Int J Mol Sci. 2023;24(2):1545.
    
    

69. 69.Ehlen JC, Brager AJ, Baggs J, Pinckney L, Gray CL, DeBruyne JP, et al. Bmal1 function in skeletal muscle regulates sleep. Elife. 2017;6:e26557.
    
    

70. 70.Haque SN, Booreddy SR, Welsh DK. Effects of BMAL1 manipulation on the brain’s master circadian clock and behavior. Yale J Biol Med. 2019;92(2):251–8.
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=31249486&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

71. 71.Hoffmann TJ, Choquet H, Yin J, Banda Y, Kvale MN, Glymour M, et al. A large multiethnic genome-wide association study of adult body mass index identifies novel loci. Genetics. 2018;210(2):499–515.
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6ODoiZ2VuZXRpY3MiO3M6NToicmVzaWQiO3M6OToiMjEwLzIvNDk5IjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjMvMDgvMTUvMjAyMy4wOC4wOS4yMzI5Mzg3OS5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 

72. 72.Peters T, Antel J, Naaresh R, Laabs BH, Föcker M, Albers N, et al. Suggestive evidence for causal effect of leptin levels on risk for anorexia nervosa: Results of a mendelian randomization study. Front Genet. 2021;12:733606.
    
    

73. 73.Schalla MA, Stengel A. The role of ghrelin in anorexia nervosa. Int J Mol Sci. 2018;19(7):2117.
    
    

74. 74.Tural U, Iosifescu DV. Adiponectin in anorexia nervosa and its modifiers: A meta-regression study. Int J Eat Disord. 2022;55(10):1279–1290
    
    

75. 75.Mariani S, Di Giorgio MR, Rossi E, Tozzi R, Contini S, Bauleo L, et al. Blood SIRT1 shows a coherent association with leptin and adiponectin in relation to the degree and distribution of adiposity: A study in obesity, normal weight and anorexia nervosa. Nutrients. 2020;12(11): 3506.
    
    

76. 76.Kennaway DJ, Varcoe TJ, Voultsios A, Boden MJ. Global loss of Bmal1 expression alters adipose tissue hormones, gene expression and glucose metabolism. PLoS One. 2013;8(6):e65255.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pone.0065255&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=23750248&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

77. 77.Laermans J, Vancleef L, Tack J, Depoortere I. Role of the clock gene Bmal1 and the gastric ghrelin-secreting cell in the circadian regulation of the ghrelin-GOAT system. Sci Rep. 2015;5:16748.
    
    

78. 78.Gooley JJ. Circadian regulation of lipid metabolism. Proceedings of the Nutrition Society. 2016;75(4):440–50.
    
    

79. 79.Chu G, Ma G, Sun J, Zhu Y, Xiang A, Yang G, et al. Leptin receptor mediates Bmal1 regulation of estrogen synthesis in granulosa cells. Animals (Basel*).* 2019;9(11):899.
    
    

80. 80.Yang DF, Huang WC, Wu CW, Huang CY, Yang YCSH, Tung YT. Acute sleep deprivation exacerbates systemic inflammation and psychiatry disorders through gut microbiota dysbiosis and disruption of circadian rhythms. Microbiological Research. 2023;268:127292.
    
    

81. 81.Hedman A, Breithaupt L, Hübel C, Thornton LM, Tillander A, Norring C, et al. Bidirectional relationship between eating disorders and autoimmune diseases. Journal of Child Psychology and Psychiatry. 2019;60(7):803–12.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/jcpp.12958&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=30178543&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 

82. 82.Zerwas S, Larsen JT, Petersen L, Thornton LM, Quaranta M, Koch SV, et al. Eating disorders, autoimmune, and autoinflammatory disease. Pediatrics. 2017;140(6).
    
    

83. 83.Galmiche M, Achamrah N, Déchelotte P, Ribet D, Breton J. Role of microbiota-gut-brain axis dysfunctions induced by infections in the onset of anorexia nervosa. Nutr Rev. 2022;80(3):381–91.
    
    

84. 84.Kyle UG, Pichard C. The Dutch Famine of 1944-1945: a pathophysiological model of long-term consequences of wasting disease. Curr Opin Clin Nutr Metab Care. 2006;9(4):388–94.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1097/01.mco.0000232898.74415.42&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=16778567&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000239101600008&link_type=ISI) 

85. 85.Clair DS, Xu M, Wang P, Yu Y, Fang Y, Zhang F, et al. Rates of adult schizophrenia following prenatal exposure to the Chinese famine of 1959-1961. JAMA. 2005;294(5):557–62.
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1001/jama.294.5.557&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=16077049&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F08%2F15%2F2023.08.09.23293879.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000230937500015&link_type=ISI) 

86. 86.Li Y, He Y, Qi L, Jaddoe VW, Feskens EJM, Yang X, et al. Exposure to the Chinese famine in early life and the risk of hyperglycemia and type 2 diabetes in adulthood. Diabetes. 2010;59(10):2400–6.
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6ODoiZGlhYmV0ZXMiO3M6NToicmVzaWQiO3M6MTA6IjU5LzEwLzI0MDAiO3M6NDoiYXRvbSI7czo1MDoiL21lZHJ4aXYvZWFybHkvMjAyMy8wOC8xNS8yMDIzLjA4LjA5LjIzMjkzODc5LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

87. 87.Bianco-Miotto T, Craig JM, Gasser YP, van Dijk SJ, Ozanne SE. Epigenetics and DOHaD: from basics to birth and beyond. Journal of Developmental Origins of Health and Disease. 2017;8(5):513–9.
    
    

88. 88.Wajahat M, Bracken CP, Orang A. Emerging functions for snoRNAs and snoRNA-derived fragments. Int J Mol Sci. 2021;22(19):10193.
    
    

89. 89.Aitken S, Semple CA. The circadian dynamics of small nucleolar RNA in the mouse liver. Journal of The Royal Society Interface. 2017;14(130):20170034.
    
    

90. 90.Zhang R, Lahens NF, Ballance HI, Hughes ME, Hogenesch JB. A circadian gene expression atlas in mammals: Implications for biology and medicine. Proceedings of the National Academy of Sciences. 2014;111(45):16219–24.
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoicG5hcyI7czo1OiJyZXNpZCI7czoxMjoiMTExLzQ1LzE2MjE5IjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjMvMDgvMTUvMjAyMy4wOC4wOS4yMzI5Mzg3OS5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=)