Accessibility / Report Error

Searching for ancient balanced polymorphisms shared between Neanderthals and Modern Humans

Lucas Henriques Viscardi Vanessa Rodrigues Paixão-Côrtes David Comas Francisco Mauro Salzano Diego Rovaris Claiton Dotto Bau Carlos Eduardo G. Amorim Maria Cátira Bortolini About the authors

Abstract

Hominin evolution is characterized by adaptive solutions often rooted in behavioral and cognitive changes. If balancing selection had an important and long-lasting impact on the evolution of these traits, it can be hypothesized that genes associated with them should carry an excess of shared polymorphisms (trans- SNPs) across recent Homo species. In this study, we investigate the role of balancing selection in human evolution using available exomes from modern (Homo sapiens) and archaic humans (H. neanderthalensis and Denisovan) for an excess of trans-SNP in two gene sets: one associated with the immune system (IMMS) and another one with behavioral system (BEHS). We identified a significant excess of trans-SNPs in IMMS (N=547), of which six of these located within genes previously associated with schizophrenia. No excess of trans-SNPs was found in BEHS, but five genes in this system harbor potential signals for balancing selection and are associated with psychiatric or neurodevelopmental disorders. Our approach evidenced recent Homo trans-SNPs that have been previously implicated in psychiatric diseases such as schizophrenia, suggesting that a genetic repertoire common to the immune and behavioral systems could have been maintained by balancing selection starting before the split between archaic and modern humans.

Keywords:
Human behavior; evolution; balancing selection; immune genes; behavioral genes

Introduction

Many of the fundamental processes at the core of complex human behavior and cognitive abilities, including sensory processing, recognition of self and others, emotions, motivation, learning, memory, attention, vocalization and speech processing, executive function, as well as neural development, may be characterized by at least some degree of heritability (Vallender, 2011Vallender EJ (2011) Comparative genetic approaches to the evolution of human brain and behavior. Am J Hum Biol 23:53-64.; Gilissen et al., 2014Gilissen C, Hehir-Kwa JY, Thung DT, van de Vorst M, van Bon BWM, Willemsen MH, Kwint M, Janssen IM, Hoischen A, Schenck A, et al. (2014) Genome sequencing identifies major causes of severe intellectual disability. Nature 511:344-347.; Vissers et al., 2015Vissers LELM, Gilissen C and Veltman JA (2015) Genetic studies in intellectual disability and related disorders. Nat Rev Genet 17:9-18.; Johnson et al., 2016Johnson MR, Shkura K, Langley SR, Delahaye-Duriez A, Srivastava P, Hill WD, Rackham OJL, Davies G, Harris SE, Moreno-Moral A, et al. (2016) Systems genetics identifies a convergent gene network for cognition and neurodevelopmental disease. Nat Neurosci 19:223-32.). Assuming that genes are partly responsible for these phenotypes, genetic variation between and within species can be expected to give rise to the large behavioral repertoire observed in nature (Bendesky and Bargmann, 2011Bendesky A and Bargmann CI (2011) Genetic contributions to behavioural diversity at the gene-environment interface. Nat Rev Genet 12:809-820.). Furthermore, the diversity of these traits can be expected to be shaped by the fundamental forces of microevolution, including genetic drift, directional natural selection (either positive or negative), and balancing selection.

While directional selection tends to reduce variability close to the selected site (Lachance and Tishkoff, 2013Lachance J and Tishkoff SA (2013) Population ghenomics of human adaptation. Annu Rev Ecol Evol Syst 44:123-143.), balancing selection results in the persistence of variation in the population or species, even in the face of loss due to drift, leading to an excess of polymorphisms with intermediate frequencies (Nielsen et al., 2009Nielsen R, Hubisz MJ, Hellmann I, Torgerson D, Andrés AM, Albrechtsen A, Gutenkunst R, Adams MD, Cargill M, Boyko A, et al. (2009) Darwinian and demographic forces affecting human protein coding genes. Genome Res 19:838-849.) and increasing genetic diversity around the site of selection (Charlesworth, 2006Charlesworth D (2006) Balancing selection and its effects on sequences in nearby genome regions. PLoS Genet 2:e64.; Fijarczyk and Babik, 2015Fijarczyk A and Babik W (2015) Detecting balancing selection in genomes: Limits and prospects. Mol Ecol 24:3529-3545.). Balancing selection can result from different processes, such as heterozygote advantage (overdominance), negative frequency-dependent selection, and heterogeneity in selective advantage across time or space – all possibly acting in changing environments requiring a fast rate of adaptation (Boon et al., 2007Boon AK, Réale D and Boutin S (2007) The interaction between personality, offspring fitness and food abundance in North American red squirrels. Ecol Lett 10:1094-1104.; Wolf et al., 2007Wolf M, van Doorn GS, Leimar O and Weissing FJ (2007) Life-history trade-offs favour the evolution of animal personalities. Nature 447:581-584.; Bendesky and Bargmann, 2011Bendesky A and Bargmann CI (2011) Genetic contributions to behavioural diversity at the gene-environment interface. Nat Rev Genet 12:809-820.; Pruitt and Riechert, 2011Pruitt JNJ and Riechert SSE (2011) How within-group behavioural variation and task efficiency enhance fitness in a social group. Proc R Soc Lond B Biol Sci 278:1209-1215.; Schaschl et al., 2015Schaschl H, Huber S, Schaefer K, Windhager S, Wallner B and Fieder M (2015) Signatures of positive selection in the cis-regulatory sequences of the human oxytocin receptor (OXTR) and arginine vasopressin receptor 1a (AVPR1A) genes. BMC Evol Biol 15:85.; Taub and Page, 2016Taub DR and Page J (2016) Molecular signatures of natural selection for polymorphic genes of the human dopaminergic and serotonergic systems: A review. Front Psychol 7:857.).

While many cases of directional selection have been reported in the literature, only a handful of examples of balancing selection have been described. This may be due to several factors. On the one hand, balancing selection may be a transient state, leaving marks so subtle that their detection may be difficult using current tests, leading to a large number of false negatives (Fijarczyk and Babik, 2015Fijarczyk A and Babik W (2015) Detecting balancing selection in genomes: Limits and prospects. Mol Ecol 24:3529-3545.). On the other hand, most methods rely on the fact that balancing selection will lead to a decreased inter-population diversity. In populations that diverged not long ago, or that are experiencing some level of admixture, however, this pattern will be found at most neutral loci, thereby leading to a large number of false positives. One way of circumventing this issue is by comparison of different species, considering ancient balancing selection, in which most loci should indicate moderate to high divergence, and thus the number of false positives is expected to be smaller.

One of the best-studied targets of balancing selection is the major histocompatibility complex (MHC), which includes many examples of long-term maintenance of trans-species polymorphisms (trans-SNPs), i.e. ancient polymorphisms that survived in derived taxa (Takahata and Nei, 1990Takahata N and Nei M (1990) Allelic genealogy under overdominant and frequency-dependent selection and polymorphism of major histocompatibility complex loci. Genetics 124:967-978.; Clark, 1997Clark AG (1997) Neutral behavior of shared polymorphism. Proc Natl Acad Sci U S A 94:7730-7734.; Grimsley et al., 1998Grimsley C, Mather KA and Ober C (1998) HLA-H: A pseudogene with increased variation due to balancing selection at neighboring loci. Mol Biol Evol 15:1581-1588.; Ségurel et al., 2013Leffler EM, Gao Z, Pfeifer S, Segurel L, Auton A, Venn O, Bowden R, Bontrop R, Wall JD, Sella G, et al. (2013) Multiple instances of ancient balancing selection shared between humans and chimpanzees. Science 339:1578-1582.; Azevedo et al., 2015Azevedo L, Serrano C, Amorim A and Cooper DN (2015) Trans-species polymorphism in humans and the great apes is generally maintained by balancing selection that modulates the host immune response. Hum Genomics 9:21.). The study of these trans-SNPs has revealed a common theme, where individuals heterozygous for genes with key roles in the immune system seem to be more effective in their defense against pathogens, while at the same time presenting only a moderate inflammatory response (Cagliani et al., 2008Cagliani R, Fumagalli M, Riva S, Pozzoli U, Comi GP, Menozzi G, Bresolin N and Sironi M. (2008). The signature of long-standing balancing selection at the human defensin beta-1 promoter. Genome Biol 9:R143.; Leffler et al., 2013Leffler EM, Gao Z, Pfeifer S, Segurel L, Auton A, Venn O, Bowden R, Bontrop R, Wall JD, Sella G, et al. (2013) Multiple instances of ancient balancing selection shared between humans and chimpanzees. Science 339:1578-1582.; Azevedo et al., 2015Azevedo L, Serrano C, Amorim A and Cooper DN (2015) Trans-species polymorphism in humans and the great apes is generally maintained by balancing selection that modulates the host immune response. Hum Genomics 9:21.; Teixeira et al., 2015Teixeira JC, De Filippo C, Weihmann A, Meneu JR, Racimo F, Dannemann M, Nickel B, Fischer A, Halbwax M, Andre C, et al. (2015) Long-term balancing selection in LAD1 maintains a missense trans-species polymorphism in humans, chimpanzees, and bonobos. Mol Biol Evol 32:1186-1196.).

Because of the expected loss of shared polymorphic sites over time due to genetic drift, polymorphisms shared between species that diverged a long time ago are rare under neutrality (Clark, 1997Clark AG (1997) Neutral behavior of shared polymorphism. Proc Natl Acad Sci U S A 94:7730-7734.). Trans-SNPs common to species separated by a relatively deep evolutionary split, and without recent admixture, are therefore probably adaptive and maintained by balancing selection. Examples of such adaptive trans-SNPs were reported by Leffler et al. (2013)Leffler EM, Gao Z, Pfeifer S, Segurel L, Auton A, Venn O, Bowden R, Bontrop R, Wall JD, Sella G, et al. (2013) Multiple instances of ancient balancing selection shared between humans and chimpanzees. Science 339:1578-1582. and Teixeira et al. (2015)Teixeira JC, De Filippo C, Weihmann A, Meneu JR, Racimo F, Dannemann M, Nickel B, Fischer A, Halbwax M, Andre C, et al. (2015) Long-term balancing selection in LAD1 maintains a missense trans-species polymorphism in humans, chimpanzees, and bonobos. Mol Biol Evol 32:1186-1196. in their comparison between humans and the Pan genus, which are thought to have diverged about 8 million years ago (Moorjani et al., 2016Moorjani P, Amorim CEG, Arndt PF and Przeworski M (2016) Variation in the molecular clock of primates. Proc Natl Acad Sci U S A 38:10607-10612.). In more recently-diverged species, the presence of trans-SNPs must be interpreted with greater caution. For instance, for humans, which have an estimated effective population size (Ne) of ~10,000 individuals, 1% of neutral trans-SNPs will be preserved in the genome even after 53,000 generations (~1,6 million years) (Clark, 1997Clark AG (1997) Neutral behavior of shared polymorphism. Proc Natl Acad Sci U S A 94:7730-7734.). Assuming that the split between Homo sapiens and Homo neanderthalensis occurred around 400,000-275,000 years ago (Endicott et al., 2010Endicott P, Ho SYW and Stringer C (2010) Using genetic evidence to evaluate four palaeoanthropological hypotheses for the timing of Neanderthal and modern human origins. J Hum Evol 59:87-95.; Prüfer et al., 2014Prüfer K, Racimo F, Patterson N, Jay F, Sankararaman S, Sawyer S, Heinze A, Renaud G, Sudmant PH, de Filippo C, et al. (2014) The complete genome sequence of a Neanderthal from the Altai Mountains. Nature 505:43-49.), some trans-SNPs occurring in these species are expected to be neutral and occur due to stochastic events. Additionally, the retention of ancestral polymorphisms can be due to introgression, the exchange of genetic material between different species due to hybridization (Fijarczyk and Babik, 2015Fijarczyk A and Babik W (2015) Detecting balancing selection in genomes: Limits and prospects. Mol Ecol 24:3529-3545.). This has been described for the hybridization between archaic humans (including H. neanderthalensis and the closely related Denisovans) and some modern human populations (Green et al., 2010Green RE, Krause J, Briggs AW, Maricic T, Stenzel U, Kircher M, Patterson N, Li H, Zhai W, Fritz MH-Y, et al. (2010). A draft sequence of the Neandertal genome. Science 328:710-722.; Reich et al., 2010Reich D, Green RE, Kircher M, Krause J, Patterson N, Durand EY, Viola B, Briggs AW, Stenzel U, Johnson PLF, et al. (2010) Genetic history of an archaic hominin group from Denisova Cave in Siberia. Nature 468:1053-1060.; Meyer et al., 2012Meyer M, Kircher M, Gansauge M-T, Li H, Racimo F, Mallick S, Schraiber JG, Jay F, Prufer K, de Filippo C, et al. (2012). A high-coverage genome sequence from an archaic Denisovan individual. Science 338:222-226.).

Despite these difficulties, the investigation of trans-SNPs in the genus Homo is an exciting research goal. Hominin evolution is characterized by adaptive solutions rooted in behavioral and cognitive changes. For instance, creative thinking promotes change from prevailing modes of thought or expression, a change that can be associated with fitness gain (Nettle, 2006Nettle D (2006) Schizotypy and mental health amongst poets, visual artists, and mathematicians. J Res Pers 40:876-890.). In addition to the benefits of change over time, there are adaptive advantages to the parallel maintenance of different behavioral strategies within a species (Sih et al., 2004Sih A, Bell A and Johnson JC (2004) Behavioral syndromes: An ecological and evolutionary overview. Trends Ecol Evol 19:372-378.; Korte et al., 2005Korte SM, Koolhaas JM, Wingfield JC and McEwen BS (2005) The Darwinian concept of stress: Benefits of allostasis and costs of allostatic load and the trade-offs in health and disease. Neurosci Biobehav Rev 29:3-38.; Cagliani et al., 2009Cagliani R, Fumagalli M, Pozzoli U, Riva S, Cereda M, Comi GP, Pattini L, Bresolin N and Sironi M (2009) A complex selection signature at the human AVPR1B gene. BMC Evol Biol 9:123.; Taub and Page, 2016Taub DR and Page J (2016) Molecular signatures of natural selection for polymorphic genes of the human dopaminergic and serotonergic systems: A review. Front Psychol 7:857.). Assuming a genetic basis for these behavioral strategies, their parallel persistence can be seen as the result of balancing selection. In support of the idea for balancing selection, there have been several reports of polymorphisms in genes with known roles in modulating complex behavior in modern human and other mammals, which have likely persisted through balancing selection (Cagliani et al., 2009Cagliani R, Fumagalli M, Pozzoli U, Riva S, Cereda M, Comi GP, Pattini L, Bresolin N and Sironi M (2009) A complex selection signature at the human AVPR1B gene. BMC Evol Biol 9:123.; Schaschl et al., 2015Schaschl H, Huber S, Schaefer K, Windhager S, Wallner B and Fieder M (2015) Signatures of positive selection in the cis-regulatory sequences of the human oxytocin receptor (OXTR) and arginine vasopressin receptor 1a (AVPR1A) genes. BMC Evol Biol 15:85.; Taub and Page, 2016Taub DR and Page J (2016) Molecular signatures of natural selection for polymorphic genes of the human dopaminergic and serotonergic systems: A review. Front Psychol 7:857.).

Expanding on this idea, it is of great interest to investigate the role of balancing selection in the evolution of hominin, including human, behavior on a greater scale. If balancing selection has indeed had an important and long-lasting impact on the evolution of behavior in hominins, it can be hypothesized that genes associated with behavior should carry an excess of trans-SNPs across hominin species. Based on this hypothesis, we investigated the role of balancing selection in the evolution of behavior in hominins by studying the pattern of trans-SNPs in genes relevant to these traits in Neanderthals and modern humans.

Importantly, as previously mentioned, many available methods used to detect balancing selection were originally designed to target balancing selection operating over more than 4Ne generations (Clark, 1997Clark AG (1997) Neutral behavior of shared polymorphism. Proc Natl Acad Sci U S A 94:7730-7734.). Due to the relatively recent split between modern humans and Neanderthals, these methods are ineffective in detecting balancing selection when studying these two species. To overcome this limitation, we implemented an approach that enables the identification of an excess of trans-SNPs in groups of genes of interest in comparison to the exome background (i.e. the null distribution), which serves as a control for demographic effects, while we added control for gene length biases, GC content (and thus indirectly mutation rate), number of polymorphisms per gene and background selection. Using this approach, we were able to identify polymorphisms shared between modern humans and Neanderthals, many of which located in genes related to immunology and a few in genes playing a potential role in behavior, including genes that may contribute to personality traits and psychiatric disorders.

Material and Methods

Defining gene sets

To find genes underlying immune and behavioral systems, we defined two target gene sets, which we named IMMS (genes related to the immune system) and BEHS (genes related to behavior). We populated these gene sets by searching the AmiGO database (http://amigo.geneontology.org/cgi-bin/amigo/browse.cgi) using GeneOntology terms directly related to immune system (for IMMS) and behavior (for BEHS; supplementary material Table S1; http://amigo.geneontology.org/cgi-bin/amigo/browse.cgi). We further added to the BEHS gene set, genes associated with autism spectrum disorder (Iossifov et al., 2014Iossifov I, O’Roak BJ, Sanders SJ, Ronemus M, Krumm N, Levy D, Stessman HA, Witherspoon K, Vives L, Patterson KE, et al. (2014). The contribution of de novo coding mutations to autism spectrum disorder. Nature 13:216-221.; Yuen et al., 2015Yuen RKC, Thiruvahindrapuram B, Merico D, Walker S, Tammimies K, Hoang N, Chrysler C, Nalpathamkalam T, Pellecchia G, Liu Y, et al. (2015) Whole-genome sequencing of quartet families with autism spectrum disorder. Nat Med 21:185-191.), schizophrenia (Carrera et al., 2009Carrera N, Sanjuán J, Moltó MD, Carracedo Á and Costas J (2009) Recent adaptive selection at MAOB and ancestral susceptibility to schizophrenia. Am J Med Genet B Neuropsychiatr Genet 150:369-374.; Li et al., 2015Li J, Cai T, Jiang Y, Chen H, He X, Chen C, Li X, Shao Q, Ran X, Li Z, et al. (2015) Genes with de novo mutations are shared by four neuropsychiatric disorders discovered from NPdenovo database. Mol Psychiatry 21:290-297.; Srinivasan et al., 2015Srinivasan S, Bettella F, Mattingsdal M, Wang Y, Witoelar A, Schork AJ, Thompson WK, Zuber V, Winsvold BS, Zwart JA, et al. (2015) Genetic markers of human Eevolution are enriched in schizophrenia. Biol Psychiatry 80:284-292.), major depression (Cagliani et al., 2009Cagliani R, Fumagalli M, Pozzoli U, Riva S, Cereda M, Comi GP, Pattini L, Bresolin N and Sironi M (2009) A complex selection signature at the human AVPR1B gene. BMC Evol Biol 9:123.), and finally the OMIM database (https://www.omim.org) was also consulted for psychopathology associated genes (i.e. schizophrenia, major depressive disorder, autism spectrum disorder, asperger syndrome, attention-deficit disorder, antisocial behavior, and obsessive-compulsive disorder) expanding our total dataset. We then excluded genes that were not available for Neanderthal exome analysis, making for a final count of 1,780 genes in IMMS and 278 in BEHS, with a total of 17,246 analyzed genes including target (IMMS or BEHS, accordingly) and control genes.

Genetic datasets

The high quality exomes of three Neanderthals were retrieved from the Max Planck Institute database (http://cdna.eva.mpg.de/neandertal/exomes/; Castellano et al., 2014Castellano S, Parra G, Sánchez-Quinto FA, Racimo F, Kuhlwilm M, Kircher M, Sawyer S, Fu Q, Heinze A, Nickel B, et al. (2014) Patterns of coding variation in the complete exomes of three Neandertals. Proc Natl Acad Sci U S A 111:6666-6671.). Modern human exome data were obtained from phase 3 of the 1000 Genomes Project (The 1000 Genomes Project Consortium, 2015The 1000 Genomes Project Consortium (2015) A global reference for human genetic variation. Nature 526:68-74.). To avoid any confounding effects due to interbreeding among archaic humans and modern Eurasians (as reported by Green et al., 2010Green RE, Krause J, Briggs AW, Maricic T, Stenzel U, Kircher M, Patterson N, Li H, Zhai W, Fritz MH-Y, et al. (2010). A draft sequence of the Neandertal genome. Science 328:710-722.; Reich et al., 2010Reich D, Green RE, Kircher M, Krause J, Patterson N, Durand EY, Viola B, Briggs AW, Stenzel U, Johnson PLF, et al. (2010) Genetic history of an archaic hominin group from Denisova Cave in Siberia. Nature 468:1053-1060.; Condemi et al., 2013Condemi S, Mounier A, Giunti P, Lari M, Caramelli D and Longo L (2013) Possible interbreeding in Late Italian Neanderthals? New data from the Mezzena jaw (Monti Lessini, Verona, Italy). PLoS One 8:e59781.) in our analysis of balancing selection, we used only Yoruba genomes, as they are assumed to have no admixture history with Neanderthals or Denisovans. We included only autosomal single nucleotide biallelic loci, therefore excluding insertions and deletions (InDels), polymorphic sites with more than two alleles, variants on the sex chromosomes and mitochondrial variants. Only loci found to be heterozygous in the Neanderthal exomes were considered for shared and non-shared polymorphisms; this observed heterozygosity at the individual level was assumed to reflect population-wide polymorphism. Regarding the Yoruba (YRI) sample of 108 individuals, we only considered polymorphisms that both ancestral and derived alleles were segregating in this population. Importantly, as these Neanderthal samples from Croatia and Spain dated to more than 40 Kya (Castellano et al., 2014Castellano S, Parra G, Sánchez-Quinto FA, Racimo F, Kuhlwilm M, Kircher M, Sawyer S, Fu Q, Heinze A, Nickel B, et al. (2014) Patterns of coding variation in the complete exomes of three Neandertals. Proc Natl Acad Sci U S A 111:6666-6671.), we do not expect that any Neanderthal polymorphism is a result of modern human introgression.

Due to the spontaneous deamination of 5-methylcytosine, methylated CpG sites are more prone to mutation than other sites, which raises the probability of allelic identity by state rather than by descent (Azevedo et al., 2015Azevedo L, Serrano C, Amorim A and Cooper DN (2015) Trans-species polymorphism in humans and the great apes is generally maintained by balancing selection that modulates the host immune response. Hum Genomics 9:21.). Because of this, we performed all analyses both including and excluding SNPs located within putatively methylated CpG sites (similarly to Leffler et al., 2013Leffler EM, Gao Z, Pfeifer S, Segurel L, Auton A, Venn O, Bowden R, Bontrop R, Wall JD, Sella G, et al. (2013) Multiple instances of ancient balancing selection shared between humans and chimpanzees. Science 339:1578-1582.).

The SIFT4G Software (Ng and Henikoff, 2003Ng PC and Henikoff S (2003) SIFT: Predicting amino acid changes that affect protein function. Nucleic Acids Res 31:3812-3814.; Vaser et al., 2016Vaser R, Adusumalli S, Leng SN, Sikic M and Ng PC (2016) SIFT missense predictions for genomes. Nat Protoc 11:1-9.) was used to classify SNPs into synonymous vs. nonsynonymous substitutions and to perform a phenotype prediction for the disruption effect of mutations, allied with known references in literature. Polymorphisms within untranslated regions (UTRs) were excluded from all further analyses.

Evaluating the excess of trans-SNPs in gene sets

We searched in each gene for polymorphisms shared between Neanderthal and Yoruba exomes (i.e. trans-SNPs). We then compared the number of trans-SNPs in each one of the two target gene sets (IMMS and IBHS) to that of the 10,000 gene sets made according to random permutations of all remaining human genes for which Neanderthal exome sequences were available (total of 17,246 genes). In doing so, we always removed genes already accounted for in the target gene set accordingly. For instance, IMMS has 1,780 genes in its dataset, therefore 10,000 random gene sets were built using 15,466 genes as control. Each of the 10,000 random gene sets consisted of as many genes as the target gene set it was simulating, namely 1,780 genes for the comparison to IMMS and 278 to BEHS. The rationale behind the construction of random gene sets, was to generate a null genomic distribution for trans-SNPs, against which each target gene set was then tested. Statistical significance was determined by assessing the deviation in the number of trans-SNPs in the target gene sets in comparison to the background genomic distribution of trans-SNPs generated with the random gene sets. The bash script was used to run this analysis (https://github.com/cegamorim/excess_transSNPs).

Because all loci in each genome were subject to the same demographic history, this approach implicitly controls for demography. However, it does not automatically control for possible effects of background selection, varying mutation rates across sites, and gene size. These effects are known to affect genetic diversity and may therefore bias our results. To control for these effects, genes in the control sets were matched, on a gene-by-gene basis, to those in the IMMS and BEHS target gene sets for background selection, gene length and GC content as follows: Background selection was measured with the B statistic developed by Mcvicker et al. (2009)McVicker G, Gordon D, Davis C and Green P (2009) Widespread genomic signatures of natural selection in hominid evolution. PLoS Genet 5: e1000471., which was computed based on the expected reduction in nucleotide diversity at a neutral site due to purifying selection at other sites, as a function of recombination rates, selected site locations, deleterious mutation rate, and the distribution of selection strengths, and indicates the expected fraction of neutral diversity that is present at a given site. A value close to 0 represents near complete removal of diversity as a result of selection, while a value close to 1 indicates that selection has had little effect on diversity (McVicker et al., 2009McVicker G, Gordon D, Davis C and Green P (2009) Widespread genomic signatures of natural selection in hominid evolution. PLoS Genet 5: e1000471.). To be matched with a target gene, the value of B for a control gene needed to be within 0.1 units of the value of B for the target gene. Gene size was measured as total exonic length, in accordance with the USCS build hg19 refGene table (https://genome.ucsc.edu/). To be matched, the length of target and control genes needed to be within 400 bp of each other. GC content was calculated considering the gene coordinates described in the refGene table of UCSC build hg19. In a first step, we used BEDTools (Quinlan and Hall 2010Quinlan AR and Hall IM (2010) BEDTools: A flexible suite of utilities for comparing genomic features. Bioinformatics 26:841-842.) to extract the coding exon sequence based on these coordinates, and then used in-house scripts to determine the GC percentage. To be matched, the GC percentages of target and control genes needed to be within 5% of each other. The criterion for thresholds applied was chosen after many trials where at least one control gene in the exome was found for at least 98% of the target genes in IMMS and BEHS datasets. Those target genes that could not be matched to at least one control gene in the exome were excluded from all further analyses in both the target and control gene sets (Table S2).

All data were handled with vcftools 0.1.13 (Danecek et al., 2011Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA, Handsaker RE, Lunter G, Marth GT, Sherry ST, et al. (2011) The variant call format and VCFtools. Bioinformatics 27:2156-2158.) and bcftools, as well as using in- house Python and bash scripts. Plots and analyses were implemented in the R environment (www.r- project.org).

Population genetics analyses

In addition to the analysis of trans-SNPs, we considered the classical population genetics statistics Tajima’s D and Fst as potential markers of balancing selection, indicated by an excess of polymorphisms with low population differentiation. While shared polymorphisms can detect long-term balancing selection, Tajima’s D highlights regions with an excess of (not necessarily shared) polymorphisms, due to balancing selection or population size change. On the other hand, Fst estimates genetic differentiation among populations. Both, Tajima’s D and Fst are of interest, since together they indicate potential targets for balancing selection over an intermediate time span, as evaluated using the interval between 0.4 Ne and 4 Ne (Fijarczyk and Babik, 2015Fijarczyk A and Babik W (2015) Detecting balancing selection in genomes: Limits and prospects. Mol Ecol 24:3529-3545.). Tajima’s D for the YRI population and intercontinental Fst scores (Global Fst) for Yoruba versus Europeans (CEU) and Asians (ASN) were obtained from the 1000 Genomes Selection Browser 1.0 (http://hsb.upf.edu/)(Pybus et al., 2014Pybus M, Dall’Olio GM, Luisi P, Uzkudun M, Carreño-Torres A, Pavlidis P, Laayouni H, Bertranpetit J and Engelken J (2014) 1000 Genomes Selection Browser 1.0: A genome browser dedicated to signatures of natural selection in modern humans. Nucleic Acids Res 42:1-7.). It should be stressed that, despite the use of just H. sapiens populations, the polymorphisms selected were those shared with Neanderthals. Negative Fst values were interpreted as 0. Furthermore, since intermediate allele frequencies are a hallmark of balancing selection (Nielsen et al., 2009Nielsen R, Hubisz MJ, Hellmann I, Torgerson D, Andrés AM, Albrechtsen A, Gutenkunst R, Adams MD, Cargill M, Boyko A, et al. (2009) Darwinian and demographic forces affecting human protein coding genes. Genome Res 19:838-849.), we retrieved the average heterozygosity and standard error for all trans-SNPs from the dbSNP database (https://www.ncbi.nlm.nih.gov/projects/SNP/). A threshold for Fst of 0.04, as used by Cagliani et al. (2013)Cagliani R, Guerini FR, Rubio-Acero R, Baglio F, Forni D, Agliardi C, Griffanti L, Fumagalli M, Pozzoli U, Riva S, et al. (2013) Long-standing balancing selection in the THBS4 gene: Influence on sex-specific brain expression and gray matter volumes in Alzheimer disease. Hum Mutat 34:743-753. to represent low values for Fst among human populations, and an average heterozygosity greater than 0.400 (as employed by Pakstis et al., 2010Pakstis AJ, Speed WC, Fang R, Hyland FCL, Furtado MR, Kidd JR and Kidd KK (2010) SNPs for a universal individual identification panel. Hum Genet 127:315-324.) were considered to flag up SNPs potentially affected by balancing selection.

All relevant SNPs identified in our analyses were queried for known associations with psychiatric disease using the GWAS catalog implemented in the UCSC Table Browser and the available literature.

Results

We retrieved 22,832 heterozygous sites from the Neanderthal exome (Castellano et al., 2014Castellano S, Parra G, Sánchez-Quinto FA, Racimo F, Kuhlwilm M, Kircher M, Sawyer S, Fu Q, Heinze A, Nickel B, et al. (2014) Patterns of coding variation in the complete exomes of three Neandertals. Proc Natl Acad Sci U S A 111:6666-6671.), which 4,117 were trans-SNPs (Table S3) found within 2,519 genes. Here trans-SNPs are defined as heterozygous sites in Neanderthals that were also polymorphic in modern humans, represented by Yoruba people, according to the 1000 Genomes phase 3 data (The 1000 Genomes Project Consortium, 2015The 1000 Genomes Project Consortium (2015) A global reference for human genetic variation. Nature 526:68-74.). We note that when loci located within CpG sites are included the numbers of such polymorphisms and genes almost doubles (Table S4), which is expected due to the high mutability of such sites (Kong et al., 2012Kong A, Frigge ML, Masson G, Besenbacher S, Sulem P, Magnusson G, Gudjonsson SA, Sigurdsson A, Jonasdottir A, Wong W, et al. (2012) Rate of de novo mutations, father’s age, and disease risk. Nature 488:471-475.), some of which will potentially not be trans-SNPs but only identical by state.

We sought to evaluate if gene sets related to the immune system (IMMS; 1,754 genes) and behavior (BEHS; 271 genes) were enriched for these trans-SNPs in comparison to the genome as a whole. To do so, we built and permutated 10,000 random sets of genes to be compared to these, by matching each gene in these two sets of genes to others in the genome, controlling for background selection, gene size, and mutation rate (see Material and Methods), since these factors are known to affect the number of polymorphisms in each gene. Implicitly we were also controlling for demography, since we were comparing target genes to others in the same genome, which were therefore subject to the same demographic history. Moreover, because there is no evidence up to date of interbreeding between archaic and modern humans before H. sapiens migrated out of Africa, we used Yoruba samples to control for the effect of archaic introgression (Green et al., 2010Green RE, Krause J, Briggs AW, Maricic T, Stenzel U, Kircher M, Patterson N, Li H, Zhai W, Fritz MH-Y, et al. (2010). A draft sequence of the Neandertal genome. Science 328:710-722.; Reich et al., 2010Reich D, Green RE, Kircher M, Krause J, Patterson N, Durand EY, Viola B, Briggs AW, Stenzel U, Johnson PLF, et al. (2010) Genetic history of an archaic hominin group from Denisova Cave in Siberia. Nature 468:1053-1060.; Meyer et al., 2012Meyer M, Kircher M, Gansauge M-T, Li H, Racimo F, Mallick S, Schraiber JG, Jay F, Prufer K, de Filippo C, et al. (2012). A high-coverage genome sequence from an archaic Denisovan individual. Science 338:222-226.). Below we describe our findings using this approach for each gene set individually (IMMS and BEHS).

Signals of balancing selection in the Immune System

We identified 547 trans-SNPs in IMMS genes. This number was significantly higher than the null distribution for trans-SNP observed in the 10,000 random permutations of genes from the control set (p-value= 0.016); Table 1, Figure 1 and Figure S1; see Methods). This pattern remains even if we exclude from the analysis the CpG transitions, which are known to present a higher mutation rate (Table S3). This pattern is borderline significant when considering only shared non- synonymous substitutions (297 polymorphisms; p-value = 0.05). These results suggest that in the genus Homo, genes underlying immune system function are more likely than non-immune-genes to maintain long- term shared polymorphisms, possibly through balancing selection. Additionally, we found that Neanderthals IMMS genes harbor more heterozygous sites (2,024 polymorphisms; shared or not with modern humans) than the random sets generated by permutation (p-value < 0.001; Figure 1, column “Polymorphisms”; Table S5). Many of these loci have SNP ID numbers (rs) and have thus been found to be at least biallelic in a modern human population. Due to the known hybridization between Neanderthals and some non-African H. sapiens populations, it is difficult to determine whether they represent polymorphisms shared since their split from the common ancestral; nevertheless, these findings reinforce that genes of the immune system maintain a high level of heterozygosity in the genus Homo.

Figure 1
- Density distribution of the average number of polymorphisms per gene observed for random sets genes (blue shade) in the Neanderthal samples matched to those included in the immune system (IMMS) and behavioral system (BEHS) target gene sets (red bars).
Table 1
Percentiles of the distribution of the mean values of polymorphism per gene in 10 000 random combinations simulating each of the proposed target gene systems (IMMS and BEHS)1.

Six of the 547 trans-SNPs shared between Neanderthals and Yoruba that we identified in IMMS (rs2240464, rs56318802, rs5899, rs118014438, rs377657111, and rs14178; Table 2) are located within genes that were previously associated with schizophrenia by the Schizophrenia Working Group of the Psychiatric Genomics Consortium (Schizophrenia Working Group of the Psychiatric Genomics Consortium, 2014Schizophrenia Working Group of the Psychiatric Genomics Consortium (2014) Biological insights from 108 schizophrenia-associated genetic loci. Nature 511:421-427.). It is noteworthy that another two of the trans-SNPs in IMMS are found in genes that have been associated with this disorder according to another study: rs28919579 is located in the CD4 gene, a locus that has been linked to schizophrenia through an imbalance of CD4 cell subtypes, and rs374886374 is located within the C4A gene, a potential MHC locus that has recently been associated with schizophrenia (Sekar et al., 2016Sekar A, Bialas AR, de Rivera H, Davis A, Hammond TR, Kamitaki N, Tooley K, Presumey J, Baum M, Van Doren V, et al. (2016) Schizophrenia risk from complex variation of complement component 4. Nature 530:177-183.). C4 is a fundamental element of the classical complement cascade pathway, which rapidly recognizes and eliminates pathogens and cellular debris. In the brain, C4A is expressed in neurons and promotes synaptic pruning, which is impaired in schizophrenic patients (Sekar et al., 2016Sekar A, Bialas AR, de Rivera H, Davis A, Hammond TR, Kamitaki N, Tooley K, Presumey J, Baum M, Van Doren V, et al. (2016) Schizophrenia risk from complex variation of complement component 4. Nature 530:177-183.). The possibility that these trans-SNPs have influenced the behavioral plasticity of Neanderthal and modern humans is speculative, potentially a lot so, but these findings suggest that at least part of the common genetic repertoire that links the IMMS with schizophrenia in modern humans has been maintained polymorphic for thousands of years. We note, however, that immune system genes are known to be highly pleiotropic (Cotsapas et al., 2011Cotsapas C, Voight BF, Rossin E, Lage K, Neale BM, Wallace C, Abecasis GR, Barrett JC, Behrens T, Cho J, et al. (2011) Pervasive sharing of genetic effects in autoimmune disease. PLoS Genet 7:e1002254.; Andreassen et al., 2015Andreassen OA, Harbo HF, Wang Y, Thompson WK, Schork AJ, Mattingsdal M, Zuber V, Bettella F, Ripke S, Kelsoe JR, et al. (2015) Genetic pleiotropy between multiple sclerosis and schizophrenia but not bipolar disorder: Differential involvement of immune- related gene loci. Mol Psychiatry 20:207-214.; Wang et al., 2015Wang Q, Yang C, Gelernter J and Zhao H (2015) Pervasive pleiotropy between psychiatric disorders and immune disorders revealed by integrative analysis of multiple GWAS. Hum Genet 134:1195-1209.), and this picture may also be true for other traits besides those related to psychiatric disorders. Still, we identified another 60 trans-SNPs located within these 108 loci associated with schizophrenia (Table 2), but that were neither part of our set for IMMS nor BEHS, suggesting that other systems may have trans-SNPs in pleiotropy with schizophrenia.

Table 2
Trans-SNPs within loci associated with schizophrenia according to the Schizophrenia Working Group of the Psychiatric Genomics Consortium (2014)Schizophrenia Working Group of the Psychiatric Genomics Consortium (2014) Biological insights from 108 schizophrenia-associated genetic loci. Nature 511:421-427..

Potential signals of balancing selection in Behavior System

No excess of trans-SNPs (shared polymorphisms) was found in BEHS genes (51 SNPs in 41 genes; Table 3; Figure 1 and Figure S1). Likewise, and in contrast to the immune system, the number of heterozygous loci found in Neanderthals, shared and non-shared with Yoruba, corresponds to that expected by chance (p-value = 0.1; Figure 1, column “Polymorphisms”). Additionally, Tajima’s D was not significant for positive values (Table 3). In this regard, significant positive values for Tajima’s D would reflect more pairwise differences than segregating sites due to the increased diversity of the region surrounding the selected site, indicating old balancing selection (Nordborg et al., 1996Nordborg M, Charlesworth B and Charlesworth D (1996) Increased levels of polymorphism surrounding selectively maintained sites in highly selfing species. Proc R Soc Lond B 263:1033-1039.). On the other hand, we found low Fst and high average heterozygosity, both important indicators of potential signals of balancing selection, in five trans-SNPs (rs11176013, rs12628, rs310617, rs438042, and rs362331 (Table 3) shared between Neanderthals and Yoruba. All of these SNPs are associated with psychiatric and neurodevelopmental disorders, including schizophrenia. For instance, rs438042, located near the intron/exon boundary of THBS4 exon 3, is associated with Alzheimer Disease and might be important for splicing, since THBS4 acts in inflammatory responses and synaptogenesis (Cagliani et al., 2013Cagliani R, Guerini FR, Rubio-Acero R, Baglio F, Forni D, Agliardi C, Griffanti L, Fumagalli M, Pozzoli U, Riva S, et al. (2013) Long-standing balancing selection in the THBS4 gene: Influence on sex-specific brain expression and gray matter volumes in Alzheimer disease. Hum Mutat 34:743-753.; Cocchi et al., 2015Cocchi E, Drago A and Serretti A (2015) Hippocampal pruning as a new theory of schizophrenia etiopathogenesis. Mol Neurobiol 53:2065-2081.). Another trans-SNP identified in the BEHS gene set, rs310617, located in the EEF1A gene, has been found to be heterozygous in the Denisova specimen. Although the specific role of rs310617 is not known, other substitutions in this gene have been associated with severe intellectual disability and epileptic encephalopathy (Nakajima et al., 2015Nakajima J, Okamoto N, Tohyama J, Kato M, Arai H, Funahashi O, Tsurusaki Y, Nakashima M, Kawashima H, Saitsu H, et al. (2015) De novo EEF1A2 mutations in patients with characteristic facial features, intellectual disability, autistic behaviors and epilepsy. Clin Genet 87:356-361.; Inui et al., 2016Inui T, Kobayashi S, Ashikari Y, Sato R, Endo W, Uematsu M, Oba H, Saitsu H, Matsumoto N, Kure S, et al. (2016) Two cases of early-onset myoclonic seizures with continuous parietal delta activity caused by EEF1A2 mutations. Brain Dev 38:520-524.).

Table 3
Trans-species polymorphisms identified in genes of the BEHS gene set.

Discussion

The evolutionary history of hominins is characterized by several notable and peculiar features. Of particular importance to the successful evolutionary trajectory of the genus Homo was the emergence of a large brain capable of sustaining complex and plastic adaptive behaviors, as well as cognitive skills. In recent years, a growing body of research has countered the notion that the Neanderthals were devoid of symbolism and presented lower cognitive abilities and less sophisticated behavior than the early H. sapiens of the Paleolithic (Akazawa et al., 1993Akazawa T, Dodo Y, Muhesen S, Abdul-Salam A, Abe Y, Kondo O and Mizoguchi Y (1993) The Neanderthal Remains from Dederiyeh Cave, Syria: Interim Report. Anthropol Sci 101:361-387.; Zilhão and Trinkaus, 2002Zilhão J and Trinkaus E (2002) Anatomie, contexte archéologique et sépulture de l’enfant Gravettien De L’Abri de Lagar Velho. Praehistoria 3:131-145.; Zilhão et al., 2010Zilhão J, Angelucci DE, Badal-García E, d’Errico F, Daniel F, Dayet L, Douka K, Higham TFG, Martínez- Sánchez MJ, et al. (2010) Symbolic use of marine shells and mineral pigments by Iberian Neandertals. Proc Natl Acad Sci U S A 107:1023-1028.; Pike et al., 2012Pike AWG, Hoffmann DL, Garcia-Diez M, Pettitt PB, Alcolea J, De Balbin R, Gonzalez-Sainz C, de las Heras C, Lasheras JA, Montes R, et al. (2012) U-series dating of Paleolithic art in 11 caves in Spain. Science 336:1409-1413.; Rendu et al., 2013Rendu W, Beauval C, Crevecoeur I, Bayle P, Balzeau A and Bismuth T (2013). Evidence supporting an intentional Neandertal burial at La Chapelle-aux-Saints. Proc Natl Acad Sci U S A 111:81-86.). This is in line with the findings of recent genetic and paleoneurological research. For instance, research by Mounier et al. (2016)Mounier A, Balzeau A, Caparros M and Grimaud-Hervé D (2016) Brain, calvarium, cladistics: A new approach to an old question, who are modern humans and Neandertals? J Hum Evol 92:22-36. and Ponce-de-León et al. (2016)Ponce de León MS, Bienvenu T, Akazawa T and Zollikofer CPE (2016) Brain development is similar in Neanderthals and modern humans. Curr Biol 26:R665-R666. revealed strong similarities between modern humans and Neanderthals in both endocranial anatomy and general brain development, while a study of 162 genes related to cognition by Paixão-Côrtes et al. (2013)Paixão-Côrtes VR, Viscardi LH, Salzano FM, Cátira Bortolini M and Hünemeier T (2013) The cognitive ability of extinct hominins: Bringing down the hierarchy using genomic evidences. Am J Hum Biol 25:702-705. identified a genetic repertoire shared between extinct archaic humans and modern humans. Assuming that the use of cognitive skills and complex behavior as an adaptive strategy represent a central element of the human evolutionary trajectory (Cagliani et al., 2009Cagliani R, Fumagalli M, Pozzoli U, Riva S, Cereda M, Comi GP, Pattini L, Bresolin N and Sironi M (2009) A complex selection signature at the human AVPR1B gene. BMC Evol Biol 9:123.; Schaschl et al., 2015Schaschl H, Huber S, Schaefer K, Windhager S, Wallner B and Fieder M (2015) Signatures of positive selection in the cis-regulatory sequences of the human oxytocin receptor (OXTR) and arginine vasopressin receptor 1a (AVPR1A) genes. BMC Evol Biol 15:85.; Taub and Page, 2016)Taub DR and Page J (2016) Molecular signatures of natural selection for polymorphic genes of the human dopaminergic and serotonergic systems: A review. Front Psychol 7:857., we sought to evaluate whether natural selection, particularly in the form of balancing selection, played any role in the evolution of genes potentially related to human behavior. Examples of a role for balancing selection in behavioral plasticity have been reported from primates (Babb et al., 2010Babb PL, Fernandez-Duque E and Schurr TG (2010) AVPR1A sequence variation in monogamous owl monkeys (Aotus azarai) and its implications for the evolution of platyrrhine social behavior. J Mol Evol 71:279-297.; Claw et al., 2010Claw KG, Tito RY, Stone AC and Verrelli BC (2010) Haplotype structure and divergence at human and chimpanzee serotonin transporter and receptor genes: implications for behavioral disorder association analyses. Mol Biol Evol 27:1518-1529.; Dobson and Brent, 2013Dobson SD and Brent LJN (2013) On the evolution of the serotonin transporter linked polymorphic region (5-HTTLPR) in primates. Front Hum Neurosci 7:588.; Goto et al., 2016Goto Y, Lee Y, Yamaguchi Y and Jas E (2016) Biological mechanisms underlying evolutionary origins of psychotic and mood disorders. Neurosci Res 111:13-24.; Taub and Page, 2016Taub DR and Page J (2016) Molecular signatures of natural selection for polymorphic genes of the human dopaminergic and serotonergic systems: A review. Front Psychol 7:857.), rodents (Lonn et al., 2017Lonn E, Koskela E, Mappes T, Mokkonen M, Sims AM and Watts PC (2017) Balancing selection maintains polymorphisms at neurogenetic loci in field experiments. Proc Natl Acad Sci U S A114:3690-3695.) and even arthropods (Fitzpatrick et al., 2007Fitzpatrick MJ, Feder E, Rower L and Sokolowski MB (2007) Maintaining a behavior polymorphism by frequency-dependent selection on a single gene. Nature 447:210-212.). To detect balanced polymorphisms in genes related to behavior, we implemented an approach based on the search for an excess of shared polymorphisms (trans-SNPs) between archaic and modern humans, in comparison to the genome as a whole.

Our analyses revealed no excess of trans-SNPs in genes known to underlie behavioral traits (captured in our BEHS gene set, see Material and Methods). These findings are consistent with the current knowledge about the genetics underlying H. sapiens brain function. Genes expressed in the brain have a large number of functions and the interactions between them are complex (giving rise to basal and specific behavioral phenotypes). Therefore, these genes are subject to functional constraint (Nielsen et al., 2009Nielsen R, Hubisz MJ, Hellmann I, Torgerson D, Andrés AM, Albrechtsen A, Gutenkunst R, Adams MD, Cargill M, Boyko A, et al. (2009) Darwinian and demographic forces affecting human protein coding genes. Genome Res 19:838-849.). Recently, Aggarwala and Voight (2016)Aggarwala V and Voight BF (2016) An expanded sequence context model broadly explains variability in polymorphism levels across the human genome. Nat Genet 48:349-355. developed the concept of genic tolerance to assess the probability of nucleotide substitution in the human genome, based on factors such as population history and selection, among others. In their analysis, genes playing a role in neurodevelopmental and psychiatric disorders were found to have a strong genic intolerance to nucleotide substitution. In the context of functional complexity and constraint, relatively few, key genetic changes can lead to larger effects on certain phenotypes, in response to a specific selective pressure, while at the same time maintaining original functions. Among the results of our analyses, five trans-SNPs located within our set of BEHS (rs11176013, rs12628, rs310617, rs438042, and rs362331; Table 3) presented high average heterozygosity and low Fst values, suggesting a homogeneous distribution of both alleles between populations. These results imply that balancing selection did not have a significant role in the evolution of genes implicated in human behavior as a whole, but may have been important for the evolution of particular genes within this set. Alternatively, our approach may not have had enough statistical power to detect the effect of balancing selection on the evolution of human behavior. That could be the case if frequency dependent selection, rather than overdominance, was the main mode of balancing selection, since our test is best suited to detect overdominance. Alternatively, the signals for balancing selection in the BEHS set may significantly pre- or postdate the evolutionary split between Neanderthals and modern humans.

Another possibility that deserves to be discussed in the light of our results comes from recent findings suggesting an association between behavioral traits and genes previously implicated in the immune response (Power et al., 2015Power RA, Steinberg S, Bjornsdottir G, Rietveld CA, Abdellaoui A, Nivard MM, Johannesson M, Galesloot TE, Hottenga JJ, Willemsen G, et al. (2015) Polygenic risk scores for schizophrenia and bipolar disorder predict creativity. Nat Neurosci 18:953-955.; Sekar et al., 2016Sekar A, Bialas AR, de Rivera H, Davis A, Hammond TR, Kamitaki N, Tooley K, Presumey J, Baum M, Van Doren V, et al. (2016) Schizophrenia risk from complex variation of complement component 4. Nature 530:177-183.). Through our inter-Homo exonic trans-SNP approach, we found that genes underlying immune function (found in the IMMS gene set) contain more ancestral polymorphisms than expected by chance in both Neanderthal and modern humans. These genes have played an important immunological defense role during speciation and migration of the genus Homo in a probable similar context of their hominin common ancestral. Our results support the idea that the variability of immune genes is both a target and an outcome of balancing selection (Grimsley et al., 1998Grimsley C, Mather KA and Ober C (1998) HLA-H: A pseudogene with increased variation due to balancing selection at neighboring loci. Mol Biol Evol 15:1581-1588.; Ségurel et al., 2013Ségurel L, Thompson EE, Flutre T, Lovstad J, Venkat A, Susan W, Moyse J, Ross S, Gamble K and Sella G (2013) Correction for Segurel et al., The ABO blood group is a trans-species polymorphism in primates. Proc Natl Acad Sci U S A 110:6607-6607.). Interestingly, and perhaps surprisingly, several studies have revealed a connection between the genetics of the immune system and human behavior. For instance, some of the strongest genetic associations found for schizophrenia at the population level involve variation in the immune system loci (Schizophrenia Working Group of the Psychiatric Genomics Consortium, 2014Schizophrenia Working Group of the Psychiatric Genomics Consortium (2014) Biological insights from 108 schizophrenia-associated genetic loci. Nature 511:421-427.; Sekar et al., 2016Sekar A, Bialas AR, de Rivera H, Davis A, Hammond TR, Kamitaki N, Tooley K, Presumey J, Baum M, Van Doren V, et al. (2016) Schizophrenia risk from complex variation of complement component 4. Nature 530:177-183.). It has been suggested that some proteins of the immune system work to promote synaptic pruning (Sekar et al., 2016Sekar A, Bialas AR, de Rivera H, Davis A, Hammond TR, Kamitaki N, Tooley K, Presumey J, Baum M, Van Doren V, et al. (2016) Schizophrenia risk from complex variation of complement component 4. Nature 530:177-183.), which is impaired in schizophrenic patients (Lips et al., 2012Lips ES, Cornelisse LN, Toonen RF, Min JL, Hultman CM, Holmans PA, O’Donovan MC, Purcell SM, Smit AB, Verhage M, et al. (2012). Functional gene group analysis identifies synaptic gene groups as risk factor for schizophrenia. Mol Psychiatry 17:996-1006.). Other mechanisms involved both in the etiology of schizophrenia and in the immune system have also been suggested, including deviant immune responses (Pandarakalam 2013Pandarakalam JP (2013) Is autoimmunity involved in the aetiology of schizophrenia? Prog Neurol Psychiatry 17:24-28.). Some of the IMMS trans-SNPs identified here have previously been found to be associated with schizophrenia in modern humans (rs2240464, rs56318802, rs5899, rs118014438, rs377657111, rs14178, Table 2). Furthermore, the trans-SNP rs374886374 is located in the gene C4A, at the MHC locus, which has recently also been associated with schizophrenia in a landmark study using several Psychiatric Genomics Consortium cohorts (Sekar et al., 2016Sekar A, Bialas AR, de Rivera H, Davis A, Hammond TR, Kamitaki N, Tooley K, Presumey J, Baum M, Van Doren V, et al. (2016) Schizophrenia risk from complex variation of complement component 4. Nature 530:177-183.; nearly 65,000 individuals; p-value < 10-8).

Schizophrenia is known to have a high heritability of around 60-80%, and interestingly, it is frequently contextualized in hypotheses that attempt to explain the evolution of modern human complex behavior (Srinivasan et al., 2015Srinivasan S, Bettella F, Mattingsdal M, Wang Y, Witoelar A, Schork AJ, Thompson WK, Zuber V, Winsvold BS, Zwart JA, et al. (2015) Genetic markers of human Eevolution are enriched in schizophrenia. Biol Psychiatry 80:284-292.). One hypothesis aiming to reconcile the relatively high prevalence (~0.5 to 1%) of this disorder across human populations with its negative effect on fitness is that balancing selection is maintaining several alleles at loci contributing to creative thinking, a trait that increases fitness. Under unfavorable circumstances, however, the same alleles are thought to increase vulnerability to psychiatric disorders, including schizophrenia (Power et al., 2015Power RA, Steinberg S, Bjornsdottir G, Rietveld CA, Abdellaoui A, Nivard MM, Johannesson M, Galesloot TE, Hottenga JJ, Willemsen G, et al. (2015) Polygenic risk scores for schizophrenia and bipolar disorder predict creativity. Nat Neurosci 18:953-955.; Srinivasan et al., 2015Srinivasan S, Bettella F, Mattingsdal M, Wang Y, Witoelar A, Schork AJ, Thompson WK, Zuber V, Winsvold BS, Zwart JA, et al. (2015) Genetic markers of human Eevolution are enriched in schizophrenia. Biol Psychiatry 80:284-292.). Our findings contribute to this hypothesis and suggest that some components of the immune genetic repertoire that were maintained polymorphic in both archaic and modern humans could have indirectly influenced the evolution of human behavior. This would represent an extraordinary case of evolutionary co-option, in which IMMS genes under balancing selection harbor ancestral adaptive polymorphisms related to the behavioral plasticity of the genus Homo. Allied to our conclusion, recent studies have contributed to unveil the physiological process of autoimmunity in cognition, being proposed as the driving force of cognitive evolution in genus Homo (Nataf, 2017Nataf S (2017) Autoimmunity as a driving force of cognitive evolution. Front Neurosci 11:582.).

Finally, a range of other molecular and biological processes certainly play an important role in the evolution of the behavioral plasticity characteristic of Homo species, such as gene regulation and epigenetic mechanisms. Moreover, beyond the role of the heterozygote advantage in maintaining these polymorphisms, other forms of natural selection (frequency-dependent, directional, etc.) at multiple levels (i.e., individual, kin, and/or group levels; Polimeni and Reiss, 2003Polimeni J and Reiss JP (2003) Evolutionary perspectives on schizophrenia. Can J Psychiatry 48:34-39.; Wilson and Hölldobler, 2005Wilson EO and Hölldobler B (2005) Eusociality: Origin and consequences. Proc Natl Acad Sci U S A 102:13367-13371.; Zhang and Perc, 2016)Zhang H and Perc M (2016) Evolution of conditional cooperation under multilevel selection. Sci Rep 6:23006., together with the unequivocal role of culture (Mesoudi, 2016Mesoudi A (2016) Cultural evolution: Integrating psychology, evolution and culture. Curr Opin Psychol 7:17-22.), have shaped and, in the case of our species, continue to shape, human evolution. A full exploration of these topics is well beyond the scope of the present study, which intends only to explore and discuss some genetic and evolutionary pieces of this complex puzzle. Future studies may help to build a more complete picture of the evolution of hominin behavior.

Acknowledgments

We thank Alex Mesoudi and two anonymous reviewers for their valuable suggestions on an earlier version of the manuscript, and Ziyue Gao for helpful discussions. Financial assistance was provided by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES). CEGA was a fellow of the Science Without Borders program from CAPES foundation (BEX 8279/11-0) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (PDE 201145/2015-4), Brazil.

References

  • Aggarwala V and Voight BF (2016) An expanded sequence context model broadly explains variability in polymorphism levels across the human genome. Nat Genet 48:349-355.
  • Akazawa T, Dodo Y, Muhesen S, Abdul-Salam A, Abe Y, Kondo O and Mizoguchi Y (1993) The Neanderthal Remains from Dederiyeh Cave, Syria: Interim Report. Anthropol Sci 101:361-387.
  • Andreassen OA, Harbo HF, Wang Y, Thompson WK, Schork AJ, Mattingsdal M, Zuber V, Bettella F, Ripke S, Kelsoe JR, et al. (2015) Genetic pleiotropy between multiple sclerosis and schizophrenia but not bipolar disorder: Differential involvement of immune- related gene loci. Mol Psychiatry 20:207-214.
  • Azevedo L, Serrano C, Amorim A and Cooper DN (2015) Trans-species polymorphism in humans and the great apes is generally maintained by balancing selection that modulates the host immune response. Hum Genomics 9:21.
  • Babb PL, Fernandez-Duque E and Schurr TG (2010) AVPR1A sequence variation in monogamous owl monkeys (Aotus azarai) and its implications for the evolution of platyrrhine social behavior. J Mol Evol 71:279-297.
  • Bauer DE, Kamran SC, Lessard S, Xu J, Fujiwara Y, Lin C, Shao Z, Canver MC, Smith EC, Pinello L, et al. (2013) An erythroid enhancer of BCL11A subject to genetic variation determines fetal hemoglobin level. Science 342:253-257.
  • Bendesky A and Bargmann CI (2011) Genetic contributions to behavioural diversity at the gene-environment interface. Nat Rev Genet 12:809-820.
  • Boon AK, Réale D and Boutin S (2007) The interaction between personality, offspring fitness and food abundance in North American red squirrels. Ecol Lett 10:1094-1104.
  • Cagliani R, Fumagalli M, Riva S, Pozzoli U, Comi GP, Menozzi G, Bresolin N and Sironi M. (2008). The signature of long-standing balancing selection at the human defensin beta-1 promoter. Genome Biol 9:R143.
  • Cagliani R, Fumagalli M, Pozzoli U, Riva S, Cereda M, Comi GP, Pattini L, Bresolin N and Sironi M (2009) A complex selection signature at the human AVPR1B gene. BMC Evol Biol 9:123.
  • Cagliani R, Guerini FR, Rubio-Acero R, Baglio F, Forni D, Agliardi C, Griffanti L, Fumagalli M, Pozzoli U, Riva S, et al. (2013) Long-standing balancing selection in the THBS4 gene: Influence on sex-specific brain expression and gray matter volumes in Alzheimer disease. Hum Mutat 34:743-753.
  • Carrera N, Sanjuán J, Moltó MD, Carracedo Á and Costas J (2009) Recent adaptive selection at MAOB and ancestral susceptibility to schizophrenia. Am J Med Genet B Neuropsychiatr Genet 150:369-374.
  • Castellano S, Parra G, Sánchez-Quinto FA, Racimo F, Kuhlwilm M, Kircher M, Sawyer S, Fu Q, Heinze A, Nickel B, et al. (2014) Patterns of coding variation in the complete exomes of three Neandertals. Proc Natl Acad Sci U S A 111:6666-6671.
  • Charlesworth D (2006) Balancing selection and its effects on sequences in nearby genome regions. PLoS Genet 2:e64.
  • Clark AG (1997) Neutral behavior of shared polymorphism. Proc Natl Acad Sci U S A 94:7730-7734.
  • Claw KG, Tito RY, Stone AC and Verrelli BC (2010) Haplotype structure and divergence at human and chimpanzee serotonin transporter and receptor genes: implications for behavioral disorder association analyses. Mol Biol Evol 27:1518-1529.
  • Cocchi E, Drago A and Serretti A (2015) Hippocampal pruning as a new theory of schizophrenia etiopathogenesis. Mol Neurobiol 53:2065-2081.
  • Condemi S, Mounier A, Giunti P, Lari M, Caramelli D and Longo L (2013) Possible interbreeding in Late Italian Neanderthals? New data from the Mezzena jaw (Monti Lessini, Verona, Italy). PLoS One 8:e59781.
  • Cotsapas C, Voight BF, Rossin E, Lage K, Neale BM, Wallace C, Abecasis GR, Barrett JC, Behrens T, Cho J, et al. (2011) Pervasive sharing of genetic effects in autoimmune disease. PLoS Genet 7:e1002254.
  • Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA, Handsaker RE, Lunter G, Marth GT, Sherry ST, et al. (2011) The variant call format and VCFtools. Bioinformatics 27:2156-2158.
  • Dobson SD and Brent LJN (2013) On the evolution of the serotonin transporter linked polymorphic region (5-HTTLPR) in primates. Front Hum Neurosci 7:588.
  • Duan G, Guo S, Zhang Y, Ying Y, Huang P, Wang Q, Zhang L and Zhang X (2015) The effect of SCN9A variation on basal pain sensitivity in the general population: an experimental study in young women. J Pain 16:971-980.
  • Endicott P, Ho SYW and Stringer C (2010) Using genetic evidence to evaluate four palaeoanthropological hypotheses for the timing of Neanderthal and modern human origins. J Hum Evol 59:87-95.
  • Fitzpatrick MJ, Feder E, Rower L and Sokolowski MB (2007) Maintaining a behavior polymorphism by frequency-dependent selection on a single gene. Nature 447:210-212.
  • Fijarczyk A and Babik W (2015) Detecting balancing selection in genomes: Limits and prospects. Mol Ecol 24:3529-3545.
  • Gilissen C, Hehir-Kwa JY, Thung DT, van de Vorst M, van Bon BWM, Willemsen MH, Kwint M, Janssen IM, Hoischen A, Schenck A, et al. (2014) Genome sequencing identifies major causes of severe intellectual disability. Nature 511:344-347.
  • Goto Y, Lee Y, Yamaguchi Y and Jas E (2016) Biological mechanisms underlying evolutionary origins of psychotic and mood disorders. Neurosci Res 111:13-24.
  • Green RE, Krause J, Briggs AW, Maricic T, Stenzel U, Kircher M, Patterson N, Li H, Zhai W, Fritz MH-Y, et al. (2010). A draft sequence of the Neandertal genome. Science 328:710-722.
  • Grimsley C, Mather KA and Ober C (1998) HLA-H: A pseudogene with increased variation due to balancing selection at neighboring loci. Mol Biol Evol 15:1581-1588.
  • Gripp KW, Stabley DL, Geller PL, Hopkins E, Stevenson DA, Carey JC and Sol-Church K (2011) Molecular confirmation of HRAS p.G12S in siblings with Costello syndrome. Am J Med Genet Part A 155:2263-2268.
  • Heid IM, Vollmert C, Kronenberg F, Huth C, Ankerst DP, Luchner A, Hinney A, Brönner G, Wichmann HE, Illig T, et al. (2008) Association of the MC4R V103I polymorphism with the metabolic syndrome: The KORA Study. Obesity 16:369-376.
  • Inui T, Kobayashi S, Ashikari Y, Sato R, Endo W, Uematsu M, Oba H, Saitsu H, Matsumoto N, Kure S, et al. (2016) Two cases of early-onset myoclonic seizures with continuous parietal delta activity caused by EEF1A2 mutations. Brain Dev 38:520-524.
  • Iossifov I, O’Roak BJ, Sanders SJ, Ronemus M, Krumm N, Levy D, Stessman HA, Witherspoon K, Vives L, Patterson KE, et al. (2014). The contribution of de novo coding mutations to autism spectrum disorder. Nature 13:216-221.
  • Johnson MR, Shkura K, Langley SR, Delahaye-Duriez A, Srivastava P, Hill WD, Rackham OJL, Davies G, Harris SE, Moreno-Moral A, et al. (2016) Systems genetics identifies a convergent gene network for cognition and neurodevelopmental disease. Nat Neurosci 19:223-32.
  • Kay C, Skotte NH, Southwell AL and Hayden MR (2014) Personalized gene silencing therapeutics for Huntington disease. Clin Genet 86:29-36.
  • Kempf L, Nicodemus KK, Kolachana B, Vakkalanka R, Verchinski BA, Egan MF, Straub RE, Mattay VA, Callicott JH, Weinberger DR, et al. (2008) Functional polymorphisms in PRODH are associated with risk and protection for schizophrenia and fronto-striatal structure and function. PLoS Genet 4:e1000252.
  • Khan ZA, Chhabra H and Mohan C (2013) Computational analysis of the coding single nucleotide polymorphisms of disrupted in schizophrenia 1 (DISC 1) gene. World J Med Sci 8:28-35.
  • Kong A, Frigge ML, Masson G, Besenbacher S, Sulem P, Magnusson G, Gudjonsson SA, Sigurdsson A, Jonasdottir A, Wong W, et al. (2012) Rate of de novo mutations, father’s age, and disease risk. Nature 488:471-475.
  • Korte SM, Koolhaas JM, Wingfield JC and McEwen BS (2005) The Darwinian concept of stress: Benefits of allostasis and costs of allostatic load and the trade-offs in health and disease. Neurosci Biobehav Rev 29:3-38.
  • Kress JA, Kühnlein P, Winter P, Ludolph AC, Kassubek J, Müller U and Sperfeld AD (2005) Novel mutation in the ALS2 gene in juvenile amyotrophic lateral sclerosis. Ann Neurol 58:800-803.
  • Lachance J and Tishkoff SA (2013) Population ghenomics of human adaptation. Annu Rev Ecol Evol Syst 44:123-143.
  • Leffler EM, Gao Z, Pfeifer S, Segurel L, Auton A, Venn O, Bowden R, Bontrop R, Wall JD, Sella G, et al. (2013) Multiple instances of ancient balancing selection shared between humans and chimpanzees. Science 339:1578-1582.
  • Li J, Cai T, Jiang Y, Chen H, He X, Chen C, Li X, Shao Q, Ran X, Li Z, et al. (2015) Genes with de novo mutations are shared by four neuropsychiatric disorders discovered from NPdenovo database. Mol Psychiatry 21:290-297.
  • Lips ES, Cornelisse LN, Toonen RF, Min JL, Hultman CM, Holmans PA, O’Donovan MC, Purcell SM, Smit AB, Verhage M, et al. (2012). Functional gene group analysis identifies synaptic gene groups as risk factor for schizophrenia. Mol Psychiatry 17:996-1006.
  • Lombardi MS, Jaspers L, Spronkmans C, Gellera C, Taroni F, Di Maria E, Donato SD and Kaemmerer WF (2009) A majority of Huntington’s disease patients may be treatable by individualized allele-specific RNA interference. Exp Neurol 217:312-319.
  • Lonn E, Koskela E, Mappes T, Mokkonen M, Sims AM and Watts PC (2017) Balancing selection maintains polymorphisms at neurogenetic loci in field experiments. Proc Natl Acad Sci U S A114:3690-3695.
  • Ludwig KU, Mattheisen M, Mühleisen TW, Roeske D, Schmäl C, Breuer R, Schulte-Körne G, Müller-Myhsok B, Nöthen MM, Hoffmann P, et al. (2009) Supporting evidence for LRRTM1 imprinting effects in schizophrenia. Mol Psychiatry 14:743-745.
  • Mata IF, Ross OA, Kachergus J, Huerta C, Ribacoba R, Moris G, Blazquez M, Guisasola LM, Salvador C, Martinez C, et al. (2006) LRRK2 mutations are a common cause of Parkinson’s disease in Spain. Eur J Neurol 13:391-394.
  • McVicker G, Gordon D, Davis C and Green P (2009) Widespread genomic signatures of natural selection in hominid evolution. PLoS Genet 5: e1000471.
  • Mesoudi A (2016) Cultural evolution: Integrating psychology, evolution and culture. Curr Opin Psychol 7:17-22.
  • Meyer M, Kircher M, Gansauge M-T, Li H, Racimo F, Mallick S, Schraiber JG, Jay F, Prufer K, de Filippo C, et al. (2012). A high-coverage genome sequence from an archaic Denisovan individual. Science 338:222-226.
  • Moorjani P, Amorim CEG, Arndt PF and Przeworski M (2016) Variation in the molecular clock of primates. Proc Natl Acad Sci U S A 38:10607-10612.
  • Mounier A, Balzeau A, Caparros M and Grimaud-Hervé D (2016) Brain, calvarium, cladistics: A new approach to an old question, who are modern humans and Neandertals? J Hum Evol 92:22-36.
  • Nakajima J, Okamoto N, Tohyama J, Kato M, Arai H, Funahashi O, Tsurusaki Y, Nakashima M, Kawashima H, Saitsu H, et al. (2015) De novo EEF1A2 mutations in patients with characteristic facial features, intellectual disability, autistic behaviors and epilepsy. Clin Genet 87:356-361.
  • Nataf S (2017) Autoimmunity as a driving force of cognitive evolution. Front Neurosci 11:582.
  • Nettle D (2006) Schizotypy and mental health amongst poets, visual artists, and mathematicians. J Res Pers 40:876-890.
  • Ng PC and Henikoff S (2003) SIFT: Predicting amino acid changes that affect protein function. Nucleic Acids Res 31:3812-3814.
  • Nielsen R, Hubisz MJ, Hellmann I, Torgerson D, Andrés AM, Albrechtsen A, Gutenkunst R, Adams MD, Cargill M, Boyko A, et al. (2009) Darwinian and demographic forces affecting human protein coding genes. Genome Res 19:838-849.
  • Nordborg M, Charlesworth B and Charlesworth D (1996) Increased levels of polymorphism surrounding selectively maintained sites in highly selfing species. Proc R Soc Lond B 263:1033-1039.
  • Ota VK, Bellucco FT, Gadelha A, Santoro ML, Noto C, Christofolini DM, Assunção IB, Yamada KM, Ribeiro-dos-Santos AK, Santos S, et al. (2014) PRODH polymorphisms, cortical volumes and thickness in schizophrenia. PLoS One 9:e87686.
  • Paixão-Côrtes VR, Viscardi LH, Salzano FM, Cátira Bortolini M and Hünemeier T (2013) The cognitive ability of extinct hominins: Bringing down the hierarchy using genomic evidences. Am J Hum Biol 25:702-705.
  • Pakstis AJ, Speed WC, Fang R, Hyland FCL, Furtado MR, Kidd JR and Kidd KK (2010) SNPs for a universal individual identification panel. Hum Genet 127:315-324.
  • Pandarakalam JP (2013) Is autoimmunity involved in the aetiology of schizophrenia? Prog Neurol Psychiatry 17:24-28.
  • Pike AWG, Hoffmann DL, Garcia-Diez M, Pettitt PB, Alcolea J, De Balbin R, Gonzalez-Sainz C, de las Heras C, Lasheras JA, Montes R, et al. (2012) U-series dating of Paleolithic art in 11 caves in Spain. Science 336:1409-1413.
  • Polimeni J and Reiss JP (2003) Evolutionary perspectives on schizophrenia. Can J Psychiatry 48:34-39.
  • Ponce de León MS, Bienvenu T, Akazawa T and Zollikofer CPE (2016) Brain development is similar in Neanderthals and modern humans. Curr Biol 26:R665-R666.
  • Power RA, Steinberg S, Bjornsdottir G, Rietveld CA, Abdellaoui A, Nivard MM, Johannesson M, Galesloot TE, Hottenga JJ, Willemsen G, et al. (2015) Polygenic risk scores for schizophrenia and bipolar disorder predict creativity. Nat Neurosci 18:953-955.
  • Prata DP, Breen G, Munro J, Sinclair M, Osborne S, Li T, Kerwin R, St Clair D and Collier DA (2006) Bipolar 1 disorder is not associated with the RGS4, PRODH, COMT and GRK3 genes. Psychiatr Genet 16:229-230.
  • Prüfer K, Racimo F, Patterson N, Jay F, Sankararaman S, Sawyer S, Heinze A, Renaud G, Sudmant PH, de Filippo C, et al. (2014) The complete genome sequence of a Neanderthal from the Altai Mountains. Nature 505:43-49.
  • Pruitt JNJ and Riechert SSE (2011) How within-group behavioural variation and task efficiency enhance fitness in a social group. Proc R Soc Lond B Biol Sci 278:1209-1215.
  • Pybus M, Dall’Olio GM, Luisi P, Uzkudun M, Carreño-Torres A, Pavlidis P, Laayouni H, Bertranpetit J and Engelken J (2014) 1000 Genomes Selection Browser 1.0: A genome browser dedicated to signatures of natural selection in modern humans. Nucleic Acids Res 42:1-7.
  • Quinlan AR and Hall IM (2010) BEDTools: A flexible suite of utilities for comparing genomic features. Bioinformatics 26:841-842.
  • Reich D, Green RE, Kircher M, Krause J, Patterson N, Durand EY, Viola B, Briggs AW, Stenzel U, Johnson PLF, et al. (2010) Genetic history of an archaic hominin group from Denisova Cave in Siberia. Nature 468:1053-1060.
  • Rendu W, Beauval C, Crevecoeur I, Bayle P, Balzeau A and Bismuth T (2013). Evidence supporting an intentional Neandertal burial at La Chapelle-aux-Saints. Proc Natl Acad Sci U S A 111:81-86.
  • Rouskas K, Paletas K, Kalogeridis A, Sarigianni M, Ioannidou-Papagiannaki E, Tsapas and Kouvatsi A (2008) Association between BBS6/MKKS gene polymorphisms, obesity and metabolic syndrome in the Greek population. Int J Obes 32:1618-1625.
  • Schaschl H, Huber S, Schaefer K, Windhager S, Wallner B and Fieder M (2015) Signatures of positive selection in the cis-regulatory sequences of the human oxytocin receptor (OXTR) and arginine vasopressin receptor 1a (AVPR1A) genes. BMC Evol Biol 15:85.
  • Schizophrenia Working Group of the Psychiatric Genomics Consortium (2014) Biological insights from 108 schizophrenia-associated genetic loci. Nature 511:421-427.
  • Ségurel L, Thompson EE, Flutre T, Lovstad J, Venkat A, Susan W, Moyse J, Ross S, Gamble K and Sella G (2013) Correction for Segurel et al., The ABO blood group is a trans-species polymorphism in primates. Proc Natl Acad Sci U S A 110:6607-6607.
  • Sekar A, Bialas AR, de Rivera H, Davis A, Hammond TR, Kamitaki N, Tooley K, Presumey J, Baum M, Van Doren V, et al. (2016) Schizophrenia risk from complex variation of complement component 4. Nature 530:177-183.
  • Shahkarami S, Galehdari H, Ahmadzadeh A, Babaahmadi M and Pedram M. (2013) The first molecular genetics analysis of individuals suffering from nephropatic cystinosis in the Southwestern Iran. Nefrologia 33:308-315.
  • Shaltout TE, Alali KA, Bushra S, Alkaseri AM, Jose ED, Al-Khainji M, Saleh R, Salama Dahir A, Shaltout H, Al-Abdullah M, et al. (2013) Significant association of close homologue of L1 gene polymorphism rs2272522 with schizophrenia in Qatar. Asia-Pacific Psychiatry 5:17-23.
  • Sih A, Bell A and Johnson JC (2004) Behavioral syndromes: An ecological and evolutionary overview. Trends Ecol Evol 19:372-378.
  • Srinivasan S, Bettella F, Mattingsdal M, Wang Y, Witoelar A, Schork AJ, Thompson WK, Zuber V, Winsvold BS, Zwart JA, et al. (2015) Genetic markers of human Eevolution are enriched in schizophrenia. Biol Psychiatry 80:284-292.
  • Takahata N and Nei M (1990) Allelic genealogy under overdominant and frequency-dependent selection and polymorphism of major histocompatibility complex loci. Genetics 124:967-978.
  • Taub DR and Page J (2016) Molecular signatures of natural selection for polymorphic genes of the human dopaminergic and serotonergic systems: A review. Front Psychol 7:857.
  • Teixeira JC, De Filippo C, Weihmann A, Meneu JR, Racimo F, Dannemann M, Nickel B, Fischer A, Halbwax M, Andre C, et al. (2015) Long-term balancing selection in LAD1 maintains a missense trans-species polymorphism in humans, chimpanzees, and bonobos. Mol Biol Evol 32:1186-1196.
  • The 1000 Genomes Project Consortium (2015) A global reference for human genetic variation. Nature 526:68-74.
  • Tomiyama H, Kokubo Y, Sasaki R, Li Y, Imamichi Y, Funayama M, Mizuno Y, Hattori N and Kuzuhara S (2008) Mutation analyses in amyotrophic lateral sclerosis/parkinsonism-dementia complex of the Kii peninsula, Japan. Mov Disord 23:2344-2348.
  • Vallender EJ (2011) Comparative genetic approaches to the evolution of human brain and behavior. Am J Hum Biol 23:53-64.
  • Vaser R, Adusumalli S, Leng SN, Sikic M and Ng PC (2016) SIFT missense predictions for genomes. Nat Protoc 11:1-9.
  • Vissers LELM, Gilissen C and Veltman JA (2015) Genetic studies in intellectual disability and related disorders. Nat Rev Genet 17:9-18.
  • Wang Q, Yang C, Gelernter J and Zhao H (2015) Pervasive pleiotropy between psychiatric disorders and immune disorders revealed by integrative analysis of multiple GWAS. Hum Genet 134:1195-1209.
  • Wilson EO and Hölldobler B (2005) Eusociality: Origin and consequences. Proc Natl Acad Sci U S A 102:13367-13371.
  • Wolf M, van Doorn GS, Leimar O and Weissing FJ (2007) Life-history trade-offs favour the evolution of animal personalities. Nature 447:581-584.
  • Yuen RKC, Thiruvahindrapuram B, Merico D, Walker S, Tammimies K, Hoang N, Chrysler C, Nalpathamkalam T, Pellecchia G, Liu Y, et al. (2015) Whole-genome sequencing of quartet families with autism spectrum disorder. Nat Med 21:185-191.
  • Zhang H and Perc M (2016) Evolution of conditional cooperation under multilevel selection. Sci Rep 6:23006.
  • Zilhão J and Trinkaus E (2002) Anatomie, contexte archéologique et sépulture de l’enfant Gravettien De L’Abri de Lagar Velho. Praehistoria 3:131-145.
  • Zilhão J, Angelucci DE, Badal-García E, d’Errico F, Daniel F, Dayet L, Douka K, Higham TFG, Martínez- Sánchez MJ, et al. (2010) Symbolic use of marine shells and mineral pigments by Iberian Neandertals. Proc Natl Acad Sci U S A 107:1023-1028.

  • Associate Editor: Carlos F.M. Menck

Publication Dates

  • Publication in this collection
    Jan-Mar 2018

History

  • Received
    02 Oct 2017
  • Accepted
    26 Nov 2017
Sociedade Brasileira de Genética Rua Cap. Adelmio Norberto da Silva, 736, 14025-670 Ribeirão Preto SP Brazil, Tel.: (55 16) 3911-4130 / Fax.: (55 16) 3621-3552 - Ribeirão Preto - SP - Brazil
E-mail: editor@gmb.org.br