Abstract
Although depression significantly impacts fitness, some hypotheses suggest that it may offer a survival benefit. However, there has been limited systematic investigation into the selection pressures acting on genes associated with depression at the genomic level. Here, we conducted comparative genomic analyses and computational molecular evolutionary analyses on 320 depression-associated genes at two levels, i.e., across the primate phylogeny (long timescale selection) and in modern human populations (recent selection). We identified seven genes under positive selection in the human lineage, and 46 genes under positive selection in modern human populations. Most positively selected variants in modern human populations were at UTR regions and non-coding exons, indicating the importance of gene expression regulation in the evolution of depression-associated genes. Positively selected genes are not only related to immune responses, but also function in reproduction and dietary adaptation. Notably, the proportion of depression-associated genes under positive selection was significantly higher than the positively selected genes at the genome-wide average level in African, East Asian, and South Asian populations. We also identified two positively selected loci that happened to be associated with depression in the South Asian population. Our study revealed that depression-associated genes are subject to varying selection pressures across different populations. We suggest that, in precision medicine—particularly in gene therapy—it is crucial to consider the specific functions of genes within distinct populations.
Introduction
Depression ranks as the leading cause of disability and seriously affects the normal life of many people (Beurel et al. 2020). According to its typical clinical features, depression has multiple negative effects on human beings. The International Classification of Diseases (ICD-10-coded) defined depression as a mental disease in which the patient suffers from lowering of mood, reduction of energy, and decrease in activity. (Capacity for enjoyment, interest, and concentration is reduced, and marked tiredness after even minimum effort is common. Sleep is usually disturbed and appetite diminished. Self-esteem and self-confidence are almost always reduced and, even in the mild form, some ideas of guilt or worthlessness are often present. The lowered mood varies little from day to day, is unresponsive to circumstances and may be accompanied by so-called “somatic” symptoms) (https://icd.who.int/browse10/2016/en#/F32). Recent studies have also reported that most suicides have been diagnosed with depression (Hawton et al. 2013). In general, depression has considerable negative consequences, including notable costs to survival and fitness.
Although depression considerably impairs fitness, a few hypotheses propose that depression can have survival benefits. Some hypotheses focus on the phenotype of depression itself, emphasizing that emotions can affect physiological reactions and behavior, playing adaptive roles in social benefits, such as to reduce costly conflict and to avoid infection (Anders et al. 2013). Other hypotheses emphasize the associations between depression and immune functions, such as the pathogen‒host defense hypothesis (Raison and Miller 2013), which proposes that depression can help energy conservation and reallocation to enhance immune responses. To better understand the evolution of depression, we first need to systematically investigate the selection pressures on depression-associated genes at the genomic level.
Natural selection is the major force driving the adaptive evolution of human beings, and selection pressures can occur at different stages during evolution. On the one hand, selection at the species level must be considered on a long timescale, i.e., millions of years and usually working on the decisive phenotypes for speciation. For example, brain development and walking upright have been reported to be under positive selection in the human lineage (Wang and Crompton 2004; Dumas et al. 2021). On the other hand, migration out of Africa in the last 100,000 years has further shaped different modern human populations under different selection pressures, including environmental changes, diversified food resources, and new pathogens (Benton et al. 2021). These recent selection pressures work on new genotypes or gene networks that are adaptable. A series of studies have reported that genes involved in local adaptation, such as anti-UV radiation (Yang et al. 2022), pathogen antagonism (Klunk et al. 2022), and dietary adaptation (Bersaglieri et al. 2004; Kothapalli et al. 2016; Ye et al. 2017; Chen et al. 2022) are under positive selection. If these genes are also involved in diseases due to genetic pleiotropic effects, then disease-associated genes could also be under positive selection. Whether this is the case in depression-associated genes remains unknown.
Growing evidence has shown that depression has a polygenic genetic basis. With the rapid development of next generation sequencing technology, researchers have started to screen depression-associated genes at the whole-genomic level using genome-wide association studies (GWAS). Hundreds of genes have been identified to be associated with depression by using large cohorts from the UK Biobank, the Million Veteran Program, 23andMe, and FinnGen (Howard et al. 2018a, 2019; Wingo et al. 2021; Levey et al. 1.2 million individuals highlight new therapeutic directions. Nat Neurosci 24:954–963" href="https://www.nature.com/articles/s41437-025-00753-1#ref-CR37" id="ref-link-section-d91260679e527">2021). Well-assembled and annotated genomes of multiple primate species are also available. These provide good resources for us to examine the selection pressures on the evolution of depression-associated genes in depth. Therefore, in this study, we investigated the positive selection on depression-associated genes retrieved from the GWAS studies mentioned above at two different levels, i.e., across the primate phylogeny (long timescale selection, i.e., millions of years) and in modern human populations (recent selection, about 100,000 years).
Materials and methods
Positive selection detection across the phylogeny
320 depression-associated genes at autosomes identified by GWAS were retrieved from literatures (Howard et al. 2018a, 2019; Wingo et al. 2021; Levey et al. 1.2 million individuals highlight new therapeutic directions. Nat Neurosci 24:954–963" href="https://www.nature.com/articles/s41437-025-00753-1#ref-CR37" id="ref-link-section-d91260679e551">2021) (Supplementary File 1). Ensembl IDs of human genes and corresponding orthologous genes of 21 primates across genomes were retrieved from the Ensembl database using BioMart (Cunningham et al. 2022). Only one-to-one orthologous genes with high confidence were analyzed. Coding genes and gff files for each species retrieved from the Ensembl database were used to extract corresponding coding sequences. The longest transcript of each gene was used for further analyses. Codon alignments were conducted using MAFFT implemented in T-coffee (Notredame et al. 2000) with default parameters. An unrooted tree retrieved from the literature (Upham et al. 2019) was used for PAML analyses (Fig. 1).
Fig. 1: Positive selection detection of depression-associated genes across the primate phylogeny under the branch-site model implemented in PAML.
figure 1
Nine lineages were tested. The number of depression-associated genes under positive selection were given on each lineage.
Full size image
The branch-site model in PAML was used to test positive selection across the phylogeny, which consider signals among both lineages and sites (Yang 1997, 2007). The value of ω represents the ratio of nonsynonymous to synonymous substitutions. If the ω >>1, it means that the nonsynonymous mutations were favored by seletion. We detect positive selection on nine primate lineages (Fig. 1). To run the branch-site model in PAML, all branches within the target lineage were labeled as foreground branches. The M0 model was used to estimate initial branch lengths (fix_blength = 2). Alignment gaps and ambiguity characters were removed (Cleandata = 1). The modified Model A (model = 2, NSsites = 2) and the Model B (fix_omega = 1, omega = 1) were compared and a likelihood ratio test (LRT) was used to detect significance (Yang 1997, 2007). Multiple test corrections were conducted using Bonferroni correction (Anisimova and Yang 2007; Gu and Xia 2019; Gu et al. 2023).
Positive selection detection in modern human populations
Haplotype-based methods (the integrated haplotype score (iHS) and the cross-population extended haplotype homozygosity (xpEHH)) which can limit demographic effects (Sabeti et al. 2002; Voight et al. 2006; Nielsen et al. 2007), as well as pairwise FST tests (Innan and Kim 2008) were used to detect positive selection across the genome in different modern human populations, retrieved from the 1000 Genomes Project (ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/20130502/). The iHS test can detect positively selected genes that have not been fixed in the population, and the xpEHH test can detect positively selected genes that are near/at fixation (Voight et al. 2006; Sabeti et al. 2007). To further limit demographic effects, we focused on macro-populations, i.e. the African population, the East Asian population, the European population, and the South Asian population. We also removed recently admixed populations and geographically adjacent populations. Populations that we used were the same as the literature (Gu et al. 2023).
The vcf files of autosomes retrieved from the 1000 Genomes Project were processed with PLINK (Purcell et al. 2007; Chang et al. 2015) and VCFtools (Danecek et al. 2011). The GRCh37/hg19 genome was used as the reference. SNPs with indels were not used. Since allele frequency and FST are highly correlated, we grouped FST values in different allele frequency bins to define the FST threshold in each bin, and the top 5% FST was used as the threshold. FST and allele frequencies were calculated with VCFtools (Danecek et al. 2011). The selscan software was used to calculate and standardize iHS and xpEHH scores for each SNP across the genome with default parameters (Szpiech and Hernandez 2014). Only biallelic SNPs with minor allele frequency ≥ 0.05 in test populations were considered. The number of SNPs with ∣iHS∣or ∣xpEHH∣ > 2 was counted in a 51-SNP window. When under positive selection, SNPs with large absolute scores of iHS or xpEHH tend to be clustered together (Voight et al. 2006). Only SNPs with ∣iHS∣or ∣xpEHH∣ > 2 in a top 1% 51-SNP window, as well as with high FST values were considered under positive selection. Given the use of genome-wide empirical distributions for standardization, no formal significance testing was conducted when employing these haplotype-based methods, as noted in the literature (Voight et al. 2006). We focused on SNPs located in exons in this study. After identifying positively selected genes across the genome in different populations, we compared this gene list with depression-associated genes we have retrieved to identify which depression-associated genes were under positive selection. GO annotations of positively selected genes were conducted with DAVID (Huang et al. 2009; Sherman et al. 2022).
Statistical analyses at the genomic level
Fisher tests were conducted for pairwise comparisons between humans and other primates to determine whether the proportion of positively selected depression-associated genes was significantly different. To determine whether the proportion of positively selected depression-associated genes was significantly larger than the proportion of positively selected genes at the genomic level (at the autosomal level) in the target branch, we randomly selected 320 genes (the same number as depression-associated genes we tested) across the genome using homologous genes retrieved for PAML analyses each time, and recorded how many of them were under positive selection in that branch. We repeated these steps 100,000 times for each branch and plotted the density distribution. Then, we checked whether the actual number of positively selected depression-associated genes was extremely high (located outside the 5% tails of the distribution).
To determine whether the proportion of positively selected depression-associated genes in the target population was significantly larger than the proportion of positively selected genes at the genomic level (at the autosomal level) in that population, we randomly selected 320 genes (the same number as depression-associated genes we tested) across the genome each time and recorded how many of them were under positive selection in that population. We repeated these steps 100,000 times for each population and plotted the density distribution. Then we checked whether the actual number of positively selected depression-associated genes was extremely high (located outside the 5% tails of the distribution).
Gene Ontology annotations
Gene Ontology biological process terms were retrieved from AmiGO (https://amigo.geneontology.org/amigo/) (Ashburner et al. 2000; Carbon et al. 2009; Aleksander et al. 2023).
Expression quantitative trait loci (eQTLs) for positively selected variants
To see whether positively selected SNPs function as eQTLs, the GTExPortal V8 (https://www.gtexportal.org/home/) was used to acquire corresponding information (Ardlie et al. 2015). The m-value (Posterior probability from MetaSoft) was used to measure the eQTL effects in multiple tissues (Han and Eskin 2011, 2012). If m > 0.9, it means that the variant does have an eQTL effect in the tissue.
Results
Depression-associated genes under positive selection across the primate phylogeny
Among the 320 depression-associated genes we tested, seven genes were under positive selection in the human lineage (Table 1, Fig. 1), including the immune response-related gene STAU1 (Pang et al. 2021), the neurodegeneration-related gene PSEN2 (Fedeli et al. 2019), the neurological-related gene ANKK1 (Hoenicka et al. 2010), the electron transfer flavoprotein dehydrogenase ETFDH (Zhang and Zhao 2022), zinc fingers and homeoboxes 3 ZHX3, the neural development-related gene PCDH9 (Chen et al. 2024), and immune-related gene LYRM4 (Wang et al. 2023).
Table 1 Depression-associated genes under positive selection in the human lineage.
Full size table
Compared to the human lineage, more depression-associated genes were identified under positive selection in other lineages. For example, 25 depression-associated genes were under positive selection in branch 2, which is the sister lineage to human beings, including Pan troglodytes and Pan paniscus. Twenty-nine depression-associated genes were under positive selection in branch 3, the ancestral lineage leading to human beings. Moreover, there were 57 depression-associated genes under positive selection in branch 4, 35 depression-associated genes under positive selection in branch 5, 96 depression-associated genes under positive selection in branch 6, 86 depression-associated genes under positive selection in branch 7, 25 depression-associated genes under positive selection in branch 8, and 18 depression-associated genes under positive selection in branch 9 (Fig. 1).
We further tested whether the proportion of positively selected depression-associated genes was significantly larger than the proportion of positively selected genes at the whole genomic level. We performed statistical significance tests in all target lineages and did not find any significantly large result. In most cases, there were no significant differences, and the proportion of positively selected depression-associated genes was even significantly less than the proportion of positively selected genes at the genomic level in certain lineages (branch 5, branch 6, branch 7 and branch 8) (Fig. 2).
Fig. 2: Statistical analyses of depression-associated genes under positive selection in different primate lineages.
figure 2
a Statistical analyses showed that the number of positively selected depression-associated genes in non-human primate lineages were significantly larger than the one in the human lineage. b The proportion of positively selected depression-associated genes was not significantly larger than the proportion of positively selected genes at the genomic level in target primate lineages. * represents p < 0.05; ** represents p < 0.01.
Full size image
Depression-associated genes under positive selection in modern human populations
Genome-wide positive selection detection identified 215 genes under positive selection in the African population, 592 genes under positive selection in the East Asian population, 632 genes under positive selection in the European population, and 567 genes under positive selection in the South Asian population. We further identified 46 depression-associated genes under positive selection in total. Among them, nine depression-associated genes were identified under positive selection by both the iHS test and the FST test, and 45 depression-associated genes were identified under positive selection by both the xpEHH test and the FST test (Table 2, Supplementary File 2). Depression-associated genes under positive selection exhibited population-specific patterns. There were eight genes (HLA-DQA1, HLA-DQB1, MGAT4C, ZNF536, SIM1, METTL9, MYRF, PSORS1C1) under positive selection in the African population, five genes (BEND4, CDH13, SOX6, POGZ, PSMB4) under positive selection in the European population, 17 genes (SDK1, CCDC92, DENND1A, USP3, C22orf26, BEND4, ASXL3, TENM2, BSN, ZNF35, MGAT4C, CTNND1, FNIP2, NRD1, INPP4B, C7orf72, FHIT) under positive selection in the East Asian population, and 25 genes (FADS1, FADS2, HLA-DQA1, HLA-DQB1, FAM120A, DENND1A, C22orf26, BEND4, FAM120AOS, ASCC3, LPIN3, POGZ, EMILIN3, CDH9, SGIP1, ZNF660, NICN1, CCDC36, CCDC71, TCAIM, MGAT4C, ZNF445, ZNF197, ZKSCAN7, PHF2) under positive selection in the South Asian population (Table 2, Supplementary File 2). FADS1 and FADS2 which are related to dietary adaptation have been previously reported to be under positive selection (Kothapalli et al. 2016; Ye et al. 2017; Chen et al. 2022).
Table 2 Depression-associated genes under positive selection in different human populations.
Full size table
Notably, the proportions of depression-associated genes under positive selection in the African population (8 out of 320) (p < 0.05), in the East Asian population (17 out of 320) (p < 0.05), and in the South Asian population (25 out of 320) (p < 0.01) were significantly larger than the proportions of positively selected genes at the genomic level in the corresponding populations (Fig. 3).
Fig. 3: Recent selection plays important roles on the adaptive evolution of depression-associated genes in modern human populations.
figure 3
a The number of positively selected depression-associated genes and the number of positively selected genes at the genomic level in each population. b Proportions of positively selected variants located in different regions (5’ UTR, 3’ UTR, coding region and non-coding exons). c The proportion of positively selected depression-associated genes in the African population, the East Asian population and the South Asian population was significantly larger than the proportion of positively selected genes at the corresponding genomic level. d left: xpEHH plot of two positively selected variants that happened to be depression-associated variants in the South Asian population compared to the European population. middle: the distribution of FST values across the genome between the South Asian population and the European population. The FST value of two positively selected variants was located at the top 5% region of the corresponding distribution (p < 0.05). right: allele frequencies of two positively selected variants in the European population and in the South Asian population.
Full size image
Expression quantitative trait loci (eQTLs) for positively selected variants
The eQTL effects of each positively selected variant at UTR regions were obtained from the GTExPortal V8. Results showed that most positively selected variants have eQTL effects across multiple tissues, and many of which were in non-brain tissues (Fig. 4).
Fig. 4: The eQTL effects of each positively selected variant in multiple tissues.
figure 4
The m-value was used to measure the eQTL effects in multiple tissues (Han and Eskin 2011, 2012). If m > 0.9, it means that the variant does have an eQTL effect in the tissue. red, m > 0.9; blue, m ≤ 0.9. Only variants with available information in the GTEx database were shown.
Full size image
Discussion
We identified seven depression-associated genes under positive selection in the human lineage. Non-human primates showed more depression-associated genes under positive selection than humans. This phenomenon is consistent with findings indicating that humans have fewer genes under positive selection compared to non-human primates (Gibbs et al. 2007; Bakewell et al. 2007). There are several possible explanations for our findings. Recent studies indicate that brain evolution occurred not only in humans but also in non-human primates, such as the Simiiformes ancestor, which displayed significant morphological changes, including rapid brain expansion (Hu et al. 2024). Primate brain evolution has been a continuous process marked by the expansion of the neocortex and the development of higher cognitive functions (Zhuang et al. 2023; Hu et al. 2024). Another explanation could be gene pleiotropy. Many positively selected genes in non-human primates are involved in functions beyond brain development, such as reproduction, metabolism, dietary adaptation, and immunity (Supplementary 3). Similar patterns have been observed in cancer-associated (Lou et al. 2014; Vicens and Posada 2018; Gu et al. 2023) and Alzheimer’s disease-associated genes (Vamathevan et al. 2008). Additionally, environmental and societal pressures differ between humans and non-human primates, leading to distinct selective pressures on the same gene functions. For example, genes related to emotional regulation may promote social bonding in non-human primates (Platt et al. 2016), whereas excessive emotional sensitivity may lead to stress and negative effects which could be detrimental in humans (Dosari et al. 2023), making such traits less advantageous.
The seven positively selected depression-associated genes we identified in the human lineage are involved in brain development (PSEN2, ANKK1, PCDH9) and immunity (STAU1 and LYRM4), as highlighted in our results. It has been demonstrated that PSEN2 knock-in mice exhibit severe depressive behavior (Yoo et al. 2024), and PCDH9 and ANKK1 have also been reported as risk genes for depression (Roetker et al. 2012; Xiao et al. 2018). The positive selection of these genes in the human lineage may attributed to their pleiotropic effects. For example, brain-related genes have long been associated with cognitive evolution in humans (Zhuang et al. 2023; Hu et al. 2024), while immunity likely played a role in responding to new environments and pathogens during human evolution (Benton et al. 2021).
Notably, among the 46 depression-associated genes under positive selection in modern human populations, we identified two genes harboring depression-associated SNPs that are also under positive selection (Fig. 3). One is located in the gene FADS1. The odds ratio (OR) for allele T on rs174546 of FADS1 was significantly greater than 1 (OR = 1.015, p = 0.000856) (Howard et al. 2018b), indicating an increased risk of depression. Despite this, the locus is under positive selection, suggesting it may provide an adaptive advantage in certain environmental or evolutionary contexts. FADS1 is known for its role in dietary adaptation in South Asian population (Kothapalli et al. 2016; Ye et al. 2017; Chen et al. 2022), supporting the biosynthesis of long-chain polyunsaturated fatty acids, which are crucial for individuals on plant-based diets. This gene’s pleiotropic trade-off could explain the positive selection of this depression-associated locus in South Asians. Another positively selected depression-associated SNP is located in the gene ZNF445, which is involved in post-fertilization methylation maintenance and imprinting disorders (Takahashi et al. 2019). Unlike the SNP in FADS1, the SNP in ZNF445 has an OR for allele T on rs2372688 of 0.98 (p = 0.000006758) (Howard et al. 2018b), suggesting a protective effect against depression. Carriers of this variant may be less prone to depression and better adapted to environmental or social conditions, enhancing their survival and reproductive success. Further research is needed to explore the detailed function of this variant. Together, these two variants offer valuable complementary evidence that disease (depression)-associated loci can undergo positive selection in specific populations.
For the additional 44 depression-associated genes detected under positive selection in different human populations, although the loci subjected to positive selection are not directly linked to depression but rather to other loci within the genes, the positive selection of these genes may have led to their wider prevalence across human populations over time. This increased frequency could provide more opportunities for the associated depression-related variants to exert their effects, whether by indirectly promoting or inhibiting depression. Most of these genes are not only involved in neuron development, but also play roles in immune response, metabolism, dietary excess, and reproduction (e.g., spermatogenesis) (see Table 2). As we know, the environment has changed a lot after migration out of Africa in the last 100,000 years, such as exposure to diverse pathogens, diversified food resources, and social culture (Benton et al. 2021). In fact, many of the depression-associated genes that underwent positive selection in both African and South Asian populations are immune-related, such as HLA-DQA1, HLA-DQB1, and MGAT4C. Given that Africa and South Asia are located in tropical regions with greater diversity of pathogens and parasites compared to other populations, these genes could be under positive selection due to their association with immune functions. Additionally, some depression-associated genes in the South Asian population are linked to diet and metabolism, such as FADS1, FADS2, LPIN3, and SGIP1. As we mentioned above, FADS1 and FADS2 are well-known for their roles in dietary adaptation in the South Asian population (Kothapalli et al. 2016; Ye et al. 2017; Chen et al. 2022). These depression-associated genes may therefore under positive selection in modern human populations due to their pleiotropic effects.
Our results also showed that most positively selected variants were located in UTR regions and non-coding exons. In recent years, there has been increasing recognition that not all exons encode proteins (Aspden et al. 2023). For instance, a study has demonstrated that multiple non-coding exons within the brain-derived neurotrophic factor (BDNF) can produce differently spliced transcripts, potentially influencing psychiatric diseases like depression (Begni et al. 2017). Currently, research on non-coding exons in relation to depression is limited. Therefore, the positively selected sites in non-coding exons we have identified here are good candidates for further study. Regarding the UTR regions, the 5’-UTR harbors various regulatory elements, including sites for ribosome recruitment, binding sites for microRNAs, and structural components crucial for governing mRNA stability, pre-mRNA splicing, and the initiation of translation (Ryczek et al. 2023). Consequently, the 5’-UTR plays a crucial role in gene expression control. While their link to depression remains relatively sparse, one study has proposed that epigenetic methylation in the 5’-UTR region could influence glucocorticoid receptor splicing, which is implicated in depression (Turner et al. 2010). For 3’-UTR regions, they are renowned for providing binding sites for small RNAs, thus regulating gene expression at the post-transcriptional level, which significantly contributes to controlling behaviors like social interactions and anxiety, relevant to depression (Narayanan and Schratt 2020). This has been underscored by various studies, such as alternative splicing of the serotonin-1A receptor (Le François et al. 2018) and the regulation of VGF protein expression, a precursor of secreted neuropeptides (Lin et al. 2021). Therefore, the positively selected SNPs identified in UTR regions and non-coding exons in our study represent promising candidates for understanding their potential roles in depression. Besides, most of the positively selected depression-associated genes had eQTL effects in multiple tissues, especially in non-brain tissues, confirming again their pleiotropic effects.
We further revealed that the proportion of positively selected depression-associated genes was so high that it was even significantly larger than the proportions of positively selected genes at the genomic level in several modern human populations, i.e., in the African population, the East Asian population and the South Asian population, but not in the European population. As shown in Table 2, many positively selected genes in non-European populations are linked to immunity, neurodevelopment, metabolism, dietary adaptation, and reproduction. The disparities between non-European and European populations may stem from several factors. For example, diets in Africa, South Asia, and East Asia are more plant-based than in Europe. Social structures in these regions are also more conservative and community-oriented. And non-European populations face greater social competition due to economic growth, population density, and higher biodiversity. However, as mentioned above, most of the positively selected loci are not directly associated with depression, but rather are other loci within depression-associated genes. Therefore, we cannot conclude whether depression is adaptive in different populations at this stage. What we can say here is that these depression-associated genes can experience different selective pressures across populations, and these pressures are closely linked to their pleiotropic effects.
It is important to note that the depression-associated genes we tested were retrieved from individuals of European descent (Howard et al. 2018a, 2019; Wingo et al. 2021; Levey et al. 1.2 million individuals highlight new therapeutic directions. Nat Neurosci 24:954–963" href="https://www.nature.com/articles/s41437-025-00753-1#ref-CR37" id="ref-link-section-d91260679e2587">2021). However, GWAS data from non-European populations are currently limited, which restricts our ability to conduct a more comprehensive analysis. Future research, when relevant data becomes available, could provide further insights. Despite these limitations, our current study provides a valuable resource and logical framework for future analyses. Moreover, our research highlights that depression-associated genes, or more broadly, disease-associated genes, can be subject to different selective pressures in different populations, and even undergo positive selection. For example, one of the risk locus in FADS1 we identified, significantly associated with depression, was under positive selection in a specific population. This suggests that in precision medicine research, it is important to take into account the specific factor of how genes exert their functions in particular populations.
Data availability
All data generated or analyzed during this study are included in this published article and its supplementary information files. Phased genetic data of the 1000 Genomes Project of modern human populations were retrieved from ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/20130502/.
References
Aleksander SA, Balhoff J, Carbon S, Cherry JM, Drabkin HJ, Ebert D et al. (2023) The Gene Ontology knowledgebase in 2023. Genetics 224:iyad031
Anders S, Tanaka M, Kinney DK (2013) Depression as an evolutionary strategy for defense against infection. Brain Behav Immun 31:9–22
PubMedGoogle Scholar
Anisimova M, Yang Z (2007) Multiple hypothesis testing to detect lineages under positive selection that affects only a few sites. Mol Biol Evol 24:1219–28
PubMedGoogle Scholar
Ardlie KG, Deluca DS, Segrè AV, Sullivan TJ, Young TR, Gelfand ET et al. (2015) The Genotype-Tissue Expression (GTEx) pilot analysis: Multitissue gene regulation in humans. Science 348:648–660
Google Scholar
Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM et al. (2000) Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet 25:25–29
PubMedPubMed CentralGoogle Scholar
Aspden JL, Wallace EWJ, Whiffin N (2023) Not all exons are protein coding: Addressing a common misconception. Cell Genomics 3:100296
PubMedPubMed CentralGoogle Scholar
Bakewell MA, Shi P, Zhang J (2007) More genes underwent positive selection in chimpanzee evolution than in human evolution. Proc Natl Acad Sci USA 104:7489–7494
PubMedPubMed CentralGoogle Scholar
Begni V, Riva MA, Cattaneo A (2017) Cellular and molecular mechanisms of the brain-derived neurotrophic factor in physiological and pathological conditions. Clin Sci 131:123–138
Google Scholar
Benton ML, Abraham A, LaBella AL, Abbot P, Rokas A, Capra JA (2021) The influence of evolutionary history on human health and disease. Nat Rev Genet 22:269–283
PubMedPubMed CentralGoogle Scholar
Bersaglieri T, Sabeti PC, Patterson N, Vanderploeg T, Schaffner SF, Drake JA et al. (2004) Genetic signatures of strong recent positive selection at the lactase gene. Am J Hum Genet 74:1111–20
PubMedPubMed CentralGoogle Scholar
Beurel E, Toups M, Nemeroff CB (2020) The Bidirectional Relationship of Depression and Inflammation: Double Trouble. Neuron 107:234–256
PubMedPubMed CentralGoogle Scholar
Carbon S, Ireland A, Mungall CJ, Shu S, Marshall B, Lewis S et al. (2009) AmiGO: online access to ontology and annotation data. Bioinformatics 25:288–289
PubMedGoogle Scholar
Chang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ (2015) Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience 4:7
PubMedPubMed CentralGoogle Scholar
Chen H, Lin R, Lu Y, Zhang R, Gao Y, He Y et al. (2022) Tracing Bai-Yue Ancestry in Aboriginal Li People on Hainan Island. Mol Biol Evol 39:msac210
PubMedPubMed CentralGoogle Scholar
Chen X, Huang Y, Huang L, Huang Z, Hao Z-Z, Xu L et al. (2024) A brain cell atlas integrating single-cell transcriptomes across human brain regions. Nat Med 30:2679–2691
PubMedPubMed CentralGoogle Scholar
Cunningham F, Allen JE, Allen J, Alvarez-Jarreta J, Amode MR, Armean IM et al. (2022) Ensembl 2022. Nucleic Acids Res 50:D988–D995
PubMedGoogle Scholar
Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA et al. (2011) The variant call format and VCFtools. Bioinformatics 27:2156–8
PubMedPubMed CentralGoogle Scholar
Dosari M, AlDayel SK, Alduraibi KM, AlTurki AA, Aljehaiman F, Alamri S et al. (2023) Prevalence of Highly Sensitive Personality and Its Relationship With Depression, and Anxiety in the Saudi General Population. Cureus 15:e49834
PubMedPubMed CentralGoogle Scholar
Dumas G, Malesys S, Bourgeron T (2021) Systematic detection of brain protein-coding genes under positive selection during primate evolution and their roles in cognition. Genome Res 31:484–496
PubMedPubMed CentralGoogle Scholar
Fedeli C, Filadi R, Rossi A, Mammucari C, Pizzo P (2019) PSEN2 (presenilin 2) mutants linked to familial Alzheimer disease impair autophagy by altering Ca(2+) homeostasis. Autophagy 15:2044–2062
PubMedPubMed CentralGoogle Scholar
Gibbs RA, Rogers J, Katze MG, Bumgarner R, Weinstock GM, Mardis ER et al. (2007) Evolutionary and biomedical insights from the rhesus macaque genome. Science 316:222–234
PubMedGoogle Scholar
Gu L, Xia C (2019) Cluster expansion of apolipoprotein D (ApoD) genes in teleost fishes. BMC Evol Biol 19:9
PubMedPubMed CentralGoogle Scholar
Gu L, Xia C, Yang S, Yang G (2023) The adaptive evolution of cancer driver genes. BMC Genomics 24:215
PubMedPubMed CentralGoogle Scholar
Han B, Eskin E (2011) Random-effects model aimed at discovering associations in meta-analysis of genome-wide association studies. Am J Hum Genet 88:586–98
PubMedPubMed CentralGoogle Scholar
Han B, Eskin E (2012) Interpreting meta-analyses of genome-wide association studies. PLoS Genet 8:e1002555
PubMedPubMed CentralGoogle Scholar
Hawton K, Casañas I, Comabella C, Haw C, Saunders K (2013) Risk factors for suicide in individuals with depression: a systematic review. J Affect Disord 147:17–28
PubMedGoogle Scholar
Hoenicka J, Quiñones-Lombraña A, España-Serrano L, Alvira-Botero X, Kremer L, Pérez-González R et al. (2010) The ANKK1 gene associated with addictions is expressed in astroglial cells and upregulated by apomorphine. Biol Psychiatry 67:3–11
PubMedGoogle Scholar
Howard DM, Adams MJ, Shirali M, Clarke T-K, Marioni RE, Davies G et al. (2018a) Genome-wide association study of depression phenotypes in UK Biobank identifies variants in excitatory synaptic pathways. Nat Commun. 9:1470
PubMedPubMed CentralGoogle Scholar
Howard DM, Adams MJ, Clarke T-K, Hafferty JD, Gibson J, Shirali M et al. (2019) Genome-wide meta-analysis of depression identifies 102 independent variants and highlights the importance of the prefrontal brain regions. Nat Neurosci 22:343–352
PubMedPubMed CentralGoogle Scholar
Howard D, Adams M, Shirali M, Clarke T-K, Marioni R, Davies G et al. (2018b) Summary statistics for three depression phenotypes in UK Biobank, [dataset]. University of Edinburgh. https://doi.org/10.7488/ds/2350
Hu B, Zhuang X-L, Zhou L, Zhang G, Cooper DN, Wu D-D (2024) Deciphering the Role of Rapidly Evolving Conserved Elements in Primate Brain Development and Exploring Their Potential Involvement in Alzheimer’s Disease. Mol Biol Evol 41:msae001
PubMedPubMed CentralGoogle Scholar
Huang DW, Sherman BT, Lempicki RA (2009) Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 4:44–57
PubMedGoogle Scholar
Innan H, Kim Y (2008) Detecting local adaptation using the joint sampling of polymorphism data in the parental and derived populations. Genetics 179:1713–20
PubMedPubMed CentralGoogle Scholar
Klunk J, Vilgalys TP, Demeure CE, Cheng X, Shiratori M, Madej J et al. (2022) Evolution of immune genes is associated with the Black Death. Nature 611:312–319
PubMedPubMed CentralGoogle Scholar
Kothapalli KSD, Ye K, Gadgil MS, Carlson SE, O’Brien KO, Zhang JY et al. (2016) Positive Selection on a Regulatory Insertion-Deletion Polymorphism in FADS2 Influences Apparent Endogenous Synthesis of Arachidonic Acid. Mol Biol Evol 33:1726–39
PubMedPubMed CentralGoogle Scholar
Le François B, Zhang L, Mahajan GJ, Stockmeier CA, Friedman E, Albert PR (2018) A Novel Alternative Splicing Mechanism That Enhances Human 5-HT1A Receptor RNA Stability Is Altered in Major Depression. J Neurosci J Soc Neurosci 38:8200–8210
Google Scholar
Levey DF, Stein MB, Wendt FR, Pathak GA, Zhou H, Aslan M et al. (2021) Bi-ancestral depression GWAS in the Million Veteran Program and meta-analysis in >1.2 million individuals highlight new therapeutic directions. Nat Neurosci 24:954–963
PubMedPubMed CentralGoogle Scholar
Lin W-J, Zhao Y, Li Z, Zheng S, Zou J-L, Warren NA et al. (2021) An increase in VGF expression through a rapid, transcription-independent, autofeedback mechanism improves cognitive function. Transl Psychiatry 11:383
PubMedPubMed CentralGoogle Scholar
Lou DI, McBee RM, Le UQ, Stone AC, Wilkerson GK, Demogines AM et al. (2014) Rapid evolution of BRCA1 and BRCA2 in humans and other primates. BMC Evol Biol 14:155
PubMedPubMed CentralGoogle Scholar
Narayanan R, Schratt G (2020) miRNA regulation of social and anxiety-related behaviour. Cell Mol Life Sci 77:4347–4364
PubMedPubMed CentralGoogle Scholar
Nielsen R, Hellmann I, Hubisz M, Bustamante C, Clark AG (2007) Recent and ongoing selection in the human genome. Nat Rev Genet 8:857–68
PubMedPubMed CentralGoogle Scholar
Notredame C, Higgins DG, Heringa J (2000) T-coffee: a novel method for fast and accurate multiple sequence alignment. J Mol Biol 302:205–217
PubMedGoogle Scholar
Pang Y, Tan G, Yang X, Lin Y, Chen Y, Zhang J et al. (2021) Iron-sulphur cluster biogenesis factor LYRM4 is a novel prognostic biomarker associated with immune infiltrates in hepatocellular carcinoma. Cancer Cell Int 21:463
PubMedPubMed CentralGoogle Scholar
Platt ML, Seyfarth RM, Cheney DL (2016) Adaptations for social cognition in the primate brain. Philos Trans R Soc Lond Ser B Biol Sci 371:20150096
Google Scholar
Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MAR, Bender D et al. (2007) PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 81:559–75
PubMedPubMed CentralGoogle Scholar
Raison CL, Miller AH (2013) The evolutionary significance of depression in Pathogen Host Defense (PATHOS-D). Mol Psychiatry 18:15–37
PubMedGoogle Scholar
Roetker NS, Yonker JA, Lee C, Chang V, Basson JJ, Roan CL et al. (2012) Multigene interactions and the prediction of depression in the Wisconsin Longitudinal Study. BMJ Open 2:e000944
PubMedPubMed CentralGoogle Scholar
Ryczek N, Łyś A, Makałowska I (2023) The Functional Meaning of 5’UTR in Protein-Coding Genes. Int J Mol Sci 24:2976
PubMedPubMed CentralGoogle Scholar
Sabeti PC, Reich DE, Higgins JM, Levine HZP, Richter DJ, Schaffner SF et al. (2002) Detecting recent positive selection in the human genome from haplotype structure. Nature 419:832–7
PubMedGoogle Scholar
Sabeti PC, Varilly P, Fry B, Lohmueller J, Hostetter E, Cotsapas C et al. (2007) Genome-wide detection and characterization of positive selection in human populations. Nature 449:913–8
PubMedPubMed CentralGoogle Scholar
Sherman BT, Hao M, Qiu J, Jiao X, Baseler MW, Lane HC et al. (2022) DAVID: a web server for functional enrichment analysis and functional annotation of gene lists (2021 update). Nucleic Acids Res 50:W216–21
PubMedPubMed CentralGoogle Scholar
Szpiech ZA, Hernandez RD (2014) selscan: an efficient multithreaded program to perform EHH-based scans for positive selection. Mol Biol Evol 31:2824–7
PubMedPubMed CentralGoogle Scholar
Takahashi N, Coluccio A, Thorball CW, Planet E, Shi H, Offner S et al. (2019) ZNF445 is a primary regulator of genomic imprinting. Genes Dev 33:49–54
PubMedPubMed CentralGoogle Scholar
Turner JD, Alt SR, Cao L, Vernocchi S, Trifonova S, Battello N et al. (2010) Transcriptional control of the glucocorticoid receptor: CpG islands, epigenetics and more. Biochem Pharm 80:1860–1868
PubMedGoogle Scholar
Upham NS, Esselstyn JA, Jetz W (2019) Inferring the mammal tree: Species-level sets of phylogenies for questions in ecology, evolution, and conservation. PLoS Biol 17:e3000494
PubMedPubMed CentralGoogle Scholar
Vamathevan JJ, Hasan S, Emes RD, Amrine-Madsen H, Rajagopalan D, Topp SD et al. (2008) The role of positive selection in determining the molecular cause of species differences in disease. BMC Evol Biol 8:273
PubMedPubMed CentralGoogle Scholar
Vicens A, Posada D (2018) Selective Pressures on Human Cancer Genes along the Evolution of Mammals. Genes 9:582
PubMedPubMed CentralGoogle Scholar
Voight BF, Kudaravalli S, Wen X, Pritchard JK (2006) A Map of Recent Positive Selection in the Human Genome (L Hurst, Ed.). PLoS Biol 4:e72
Wang HY, Xie Y, Du H, Luo B, Li Z (2023) High LYRM4-AS1 predicts poor prognosis in patients with glioma and correlates with immune infiltration. PeerJ 11:e16104
PubMedPubMed CentralGoogle Scholar
Wang WJ, Crompton RH (2004) Analysis of the human and ape foot during bipedal standing with implications for the evolution of the foot. J Biomech 37:1831–6
PubMedGoogle Scholar
Wingo TS, Liu Y, Gerasimov ES, Gockley J, Logsdon BA, Duong DM et al. (2021) Brain proteome-wide association study implicates novel proteins in depression pathogenesis. Nat Neurosci 24:810–817
PubMedPubMed CentralGoogle Scholar
Xiao X, Zheng F, Chang H, Ma Y, Yao Y-G, Luo X-J et al. (2018) The Gene Encoding Protocadherin 9 (PCDH9), a Novel Risk Factor for Major Depressive Disorder. Neuropsychopharmacol Publ Am Coll Neuropsychopharmacol 43:1128–1137
Google Scholar
Yang Z (1997) PAML: a program package for phylogenetic analysis by maximum likelihood. Comput Appl Biosci 13:555–6
PubMedGoogle Scholar
Yang Z (2007) PAML 4: phylogenetic analysis by maximum likelihood. Mol Biol Evol 24:1586–91
PubMedGoogle Scholar
Yang Z, Bai C, Pu Y, Kong Q, Guo Y, Ouzhuluobu L et al. (2022) Genetic adaptation of skin pigmentation in highland Tibetans. Proc Natl Acad Sci USA 119:e2200421119
PubMedPubMed CentralGoogle Scholar
Ye K, Gao F, Wang D, Bar-Yosef O, Keinan A (2017) Dietary adaptation of FADS genes in Europe varied across time and geography. Nat Ecol Evol 1:167
PubMedPubMed CentralGoogle Scholar
Yoo SS, Lee DW, Ham HJ, Yeo IJ, Chang JY, Yun J et al. (2024) Presenilin-2 knock-In mice show severe depressive behavior via DVL3 downregulation. CNS Neurosci Ther 30:e14370
PubMedGoogle Scholar
Zhang B, Zhao Y (2022) Novel variant of ETFDH leading to multiple acyl-CoA dehydrogenase deficiency by promoting protein degradation via ubiquitin proteasome pathway. Clin Chim Acta 530:104–112
PubMedGoogle Scholar
Zhuang X-L, Zhang J-J, Shao Y, Ye Y, Chen C-Y, Zhou L et al. (2023) Integrative Omics Reveals Rapidly Evolving Regulatory Sequences Driving Primate Brain Evolution. Mol Biol Evol 40:msad173
PubMedPubMed CentralGoogle Scholar
Download references
Acknowledgements
This work was supported by Fundamental Research Funds for the Central Universities (20lgpy109) to LG, and by the Scientific Research Team Project of the College of Life Sciences, Beijing Normal University in 2024 to CX. Data computation was supported by National Supercomputer Center in Guangzhou, China.
Author information
Authors and Affiliations
Sun Yat-sen University Cancer Center, Guangzhou, Guangdong, 510060, China
Shiyu Yang
The Affiliated Brain Hospital, Guangzhou Medical University, Guangzhou, Guangdong, 510180, China
Shiyu Yang
State Key Laboratory for Biocontrol, School of Life Sciences, Sun Yat-sen University, Guangzhou, Guangdong, 510275, China
Chenqing Zheng & Langyu Gu
Ministry of Education Key Laboratory for Biodiversity and Ecological Engineering, College of Life Sciences, Beijing Normal University, Beijing, 100875, China
Canwei Xia
Department of Pathology, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, 510080, China
Jihui Kang
Authors
Shiyu Yang
View author publications
You can also search for this author in PubMedGoogle Scholar
2. Chenqing Zheng
View author publications
You can also search for this author in PubMedGoogle Scholar
3. Canwei Xia
View author publications
You can also search for this author in PubMedGoogle Scholar
4. Jihui Kang
View author publications
You can also search for this author in PubMedGoogle Scholar
5. Langyu Gu
View author publications
You can also search for this author in PubMedGoogle Scholar
Contributions
SY and LG designed the study. SY and LG conducted positive selection detection in modern human populations. SY, CZ and LG conducted positive selection detection across the phylogeny. SY, CX and LG conducted statistical analyses. JK conducted data curation. SY and LG wrote the manuscript and all authors commented on it.
Corresponding author
Correspondence to Langyu Gu.
Ethics declarations
Conflict of interest
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Associate editor: Sara Goodacre.
Supplementary information
Supplementary File 1
Supplementary File 2
Supplementary File 3
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Reprints and permissions
About this article
Check for updates. Verify currency and authenticity via CrossMark
Cite this article
Yang, S., Zheng, C., Xia, C. et al. Detection of positive selection on depression-associated genes. Heredity (2025). https://doi.org/10.1038/s41437-025-00753-1
Download citation
Received:08 November 2024
Accepted:24 February 2025
Published:13 March 2025
DOI:https://doi.org/10.1038/s41437-025-00753-1
Share this article
Anyone you share the following link with will be able to read this content:
Get shareable link
Sorry, a shareable link is not currently available for this article.
Copy to clipboard
Provided by the Springer Nature SharedIt content-sharing initiative