https://orcid.org/0000-0001-9303-2746
Figures
Abstract
Although genetic factors contribute to behavioral variation in working dogs, the underlying molecular determinants remain poorly understood. The sortilin-related VPS10 domain–containing receptor 1 (SORCS1) gene is highly expressed in the central nervous system and has been implicated in neuronal signaling and synaptic regulation. A preliminary genome-wide association study suggested an association between SORCS1 variants and guide dog training outcomes. However, this result relied on only 28 dogs and lacked sufficient statistical power. In this study, we expanded the sample size and evaluated the association between SORCS1 polymorphisms and behavioral suitability in 160 dogs, using pass/fail training outcomes as behavioral trait indicators. We initially validated 12 candidate SNPs within SORCS1 using Sanger sequencing in 64 dogs, revealing significant genotype-dependent differences for eight loci located in predicted intron 3. A tagging SNP (rs23402730, C > T) was subsequently genotyped in 160 dogs (65 successful and 95 unsuccessful). The strongest association was detected under a recessive model (CC + CT vs. TT), yielding P = 4.7 × 10 ⁻ ⁵ and an odds ratio (OR) of 4.11 (95% CI: 2.07–8.18). As allele frequencies and genetic relatedness varied among dogs, we further evaluated the association using generalized linear mixed-effects models adjusting for sex, breed, and family structure. In the full dataset (n = 159), the additive genotype remained statistically significant (β = 2.20, P = 4.5 × 10 ⁻ 4; OR = 8.98, 95% CI: 2.64–30.59). Moreover, we confirmed the association in a genetically homogeneous subset of 114 Labrador retrievers, independent of breed effects (P = 0.029; OR = 4.71, 95% CI: 1.17–18.99). These results suggest that variation in SORCS1 is associated with behavioral suitability in dogs, based on an expanded cohort with guide dog training outcomes. As behavior is polygenic, SORCS1 represents one contributing locus that may be informative for future genetic studies aimed at elucidating the molecular basis of canine behavior.
Citation: Omi T, Udagawa C, Kato Y, Kawakami S, Uno Y, Ochiai K, et al. (2026) Identification of SORCS1 as a candidate gene associated with canine behavioral traits: Insights from guide dog training outcomes. PLoS One 21(2): e0342346. https://doi.org/10.1371/journal.pone.0342346
Editor: Tofazzal Md Rakib, Chattogram Veterinary and Animal Sciences University, BANGLADESH
Received: October 13, 2025; Accepted: January 21, 2026; Published: February 17, 2026
Copyright: © 2026 Omi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: T.O. and J.A. hold a patent related to a part of this work. The authors declare no other competing interests.
Introduction
Genetic diversity and molecular markers are widely used to characterize dog populations and detect gene polymorphisms underlying variation in physiological and behavioral traits [1–6]. Several genes have been implicated in behavioral regulation in dogs, particularly those involved in dopaminergic, serotonergic, and neuropeptide pathways that influence impulsivity, sociability, and stress responses. For example, genes involved in neurotransmitter pathways, such as the dopamine receptors (DRD1, DRD4) [6–9], tyrosine hydroxylase (TH) [9,10], dopamine transporter (DAT/SLC6A3) [11], glutamate transporter (SLC1A2) [12], and serotonin transporter (SLC6A4/5-HTT) [13,14], have been linked to behavior. Other associated genes include catechol-O-methyltransferase (COMT) [15,16], serotonin receptor subtypes (HTR1D, HTR2C, 5HT1A, 5HT1B, and 5HT1E2A) [14,15], androgen receptor (AR) [17], oxytocin receptor (OXTR) [18–22], monoamine oxidase B (MAOB) [23], and stathmin 1 (STMN1) [24].
In particular, polymorphism in the tyrosine hydroxylase (TH) gene is associated with activity-impulsivity in German Shepherd Dogs [10], while OXTR variation has been linked to affiliative and human-directed social behavior in dogs [18]. These examples illustrate that genetic variation in neural signaling pathways could influence behavioral traits.
The sortilin-related Vps10p-domain–containing receptor 1 (SORCS1) is a biologically plausible candidate gene for behavioral regulation. SORCS1 is highly expressed in the central nervous system and plays critical roles in intracellular sorting and trafficking of synaptic receptors, including AMPA receptor complexes [25–27]. In addition, members of the Vps10p-domain receptor family have been implicated in synaptic plasticity and neurodegenerative disorders [28–33], and genomic analyses in foxes have identified SORCS1 as a candidate gene associated with tame behavior [34]. These observations provide a biological rationale for examining SORCS1 in relation to behavioral traits in dogs.
Guide dog training programs face persistent challenges worldwide due to low behavioral suitability and limited training success rates. Guide dog training outcomes reflect behavioral suitability because they encompass multiple cognitive and emotional dimensions, including learning aptitude, task persistence, stress tolerance, and social responsiveness in real operational environments [35–38]. Previous studies have applied training outcomes as objective behavioral indicators in working dogs as they capture stable behavioral traits relevant to professional performance [39–41]. Identifying genetic markers that predict success would improve selection efficiency and mitigate guide dog shortages [15,38,39].
Our preliminary genome-wide association study suggested that SORCS1 may influence behavioral suitability in guide dogs. However, as the preliminary analysis was based on only 28 dogs, the sample size was extremely small, making the determination of whether the observed association represented a true biological signal or a false positive difficult. To clarify this uncertainty, the present study increases the sample size and conducts a focused candidate-gene analysis to evaluate the association of SORCS1 variants with behavioral traits relevant to guide dog outcomes.
Methods
Dogs
A total of 160 blood samples provided by the Kyushu Guide Dog Association (Itoshima, Fukuoka, Japan) were used in this study (S1 Table, DNA Panels 1 + 2 + 3). The successful group consisted of 65 dogs, including 12 active guide dogs, 17 retired guide dogs, and 36 dogs that had successfully completed guide dog training. Conversely, the unsuccessful group comprised 95 dogs that failed guide dog training. Dogs that withdrew from training for reasons unrelated to behavioral suitability (e.g., illness or injury) were excluded from the unsuccessful group. The assessment criteria for guide dog suitability, as employed by the Kyushu Guide Dog Association, encompass a wide range of behavioral characteristics, including aggression, sensitivity, excitability, willingness, self-interest, suspicion, anxiety, adaptability, distractibility, dominance, and maturity. Such multidimensional evaluations highlight the complexity of behavioral traits underlying guide dog performance. Classification criteria for “successful” and “unsuccessful” training outcomes follow operational performance standards used by certified guide dog instructors and are generally based on procedures described in the Guide Dog Instructor Training Textbook published by the Japan Guide Dog Facilities Association [42]. In addition to the guide dog samples, genomic DNA samples from 220 dogs were used to examine single nucleotide polymorphism (SNP) distributions across different breeds. These samples were collected from six breeds commonly kept in Japan: Labrador Retrievers (n = 50), Golden Retrievers (n = 33), Chihuahuas (n = 35), French Bulldogs (n = 36), Toy Poodles (n = 36), and Shiba Inus (n = 30). All samples were collected, extracted, and stored in our laboratory as part of ongoing genetic studies.
Genomic DNA was extracted from peripheral blood using the QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany). DNA concentration and purity were assessed using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). All sample collection procedures were performed by licensed veterinarians from the Kyushu Guide Dog Association or the Veterinary Medical Teaching Hospital, Nippon Veterinary and Life Science University (NVLU). Written informed consent was obtained from all dog owners prior to sample collection.
Validation of 12 SNPs in the canine SORCS1 gene by Sanger sequencing
Twelve SNPs showing the strongest association signals within the SORCS1 gene on chromosome 28 (S1 Fig) were selected based on a preliminary GWAS of 28 dogs (14 successful and 14 unsuccessful) performed using the CanineHD BeadChip 170K SNP array (Illumina, San Diego, CA, USA). The genomic positions of these SNPs were initially annotated based on the CanFam2.0 assembly. To reflect updates in genome annotation, the positions were re-evaluated and are shown according to the more recent CanFam3.1 assembly in Table 1. Corresponding rsIDs were identified using the Genome Data Viewer (Canis lupus familiaris, CanFam3.1), and these data are summarized in Table 1. For four SNPs, discrepancies in the designation of major and minor alleles were observed between the SNP array annotation and dbSNP records; in the present study, alleles were reported based on the rsID information. To validate these candidate SNPs within the SORCS1 gene, all 12 loci were genotyped by Sanger sequencing of PCR-amplified products. Genomic DNA was obtained from 32 successful and 32 unsuccessful guide dogs (S1 Table, DNA Panels 1 + 2). Specific regions encompassing each SNP within SORCS1 were amplified by PCR using primer sets designed for each locus (S2 Table). For consistency, subsequent references to SNPs use rs numbers.
PCR reactions were performed in a 25 μL volume using FastStart Taq DNA Polymerase (Roche, Mannheim, Germany) according to the manufacturer’s instructions. Thermal cycling was conducted with a Veriti 96-well Thermal Cycler (Applied Biosystems, Foster City, CA, USA). Amplification products were confirmed by electrophoresis on 2% agarose gels stained with ethidium bromide (Nippon Gene Co., Ltd., Toyama, Japan).
PCR products were purified using the High Pure PCR Product Purification Kit (Roche, Mannheim, Germany) and subjected to direct sequencing using the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems). Sequencing reactions were purified with the BigDye XTerminator Purification Kit (Applied Biosystems) and analyzed on an ABI 3730 Genetic Analyzer (Applied Biosystems).
Validation of a tagging SNP in the canine SORCS1 gene
Among the 12 SNPs identified within the canine SORCS1 gene, rs23402730 (BICF2G630271656) showed one of the strongest association signals and was located within a region of high linkage disequilibrium with neighboring SNPs. Based on both its strong association signal and its representativeness within this LD block, rs23402730 was selected as a tagging SNP for subsequent analyses. Genotyping of this SNP was performed by Sanger sequencing of PCR-amplified products or by real-time PCR using a TaqMan probe specifically designed for this study (S2 Table). We collected additional genomic DNA samples from 33 successful and 63 unsuccessful guide dogs (S1 Table 1, DNA Panel 3). Finally, genotyping of the SORCS1 SNP rs23402730 (C > T) locus was carried out using genomic DNA from 160 dogs (S1 Table, DNA Panels 1 + 2 + 3), consisting of 65 successful and 95 unsuccessful guide dogs.
Sequencing of the predicted coding region of the canine SORCS1 gene using genomic DNA
To investigate coding region variants potentially associated with guide dog suitability, the predicted 27 exons of the canine SORCS1 gene were amplified from the genomic DNA of two dogs homozygous for the informative CC genotype at the SORCS1 SNP rs23402730 locus. PCR primers (S2 Table) were designed based on the canine SORCS1 gene sequence (NCBI Reference Sequence: NC_006610.3), located on chromosome 28 of the Canis lupus familiaris Boxer genome (CanFam3.1, whole-genome shotgun sequence).
Expression analysis of the SORCS1 gene in canine tissues
To investigate the expression of the SORCS1 gene, reverse transcription PCR (RT-PCR) was performed using cDNA synthesized from various canine tissues. cDNA samples were obtained from Zyagen (San Diego, CA, USA) and BioChain Institute, Inc. (Newark, CA, USA). The canine SORCS1 cDNA, corresponding to the predicted exons 3–5 of the SORCS1 gene, was amplified by PCR. The canine glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene was used as an internal control [43]. The PCR products were electrophoresed on 2% agarose gels, stained with ethidium bromide, and visualized under ultraviolet illumination.
Statistical analyses
Statistical significance was evaluated using Fisher’s exact test (two-tailed), with P < 0.05 considered statistically significant. Odds ratios (ORs) and 95% confidence intervals (CI) were calculated using 2 × 2 contingency tables. Linkage disequilibrium (LD) was assessed by calculating r² values. These analyses were performed using SNPAlyze (Dynacom, Mobara, Japan) and BellCurve for Excel v4.09 (Social Survey Research Information Co., Ltd., Tokyo, Japan).
To further adjust for sex, breed, and family effects, generalized linear mixed-effects models (GLMMs) with a binomial distribution and logit link were used to test the association between SORCS1 genotype and training outcome. Genotype was coded additively, sex and breed were included as fixed effects, and family was included as a random intercept. Models were fitted using the lme4 package in R (version 4.5.2), and results are reported as ORs with 95% CI. A secondary analysis was conducted in Labrador retrievers only.
Results
Association between SORCS1 variants and guide dog training outcomes
A previous preliminary genome-wide association analysis conducted in 2015 using 28 dogs (14 successful and 14 unsuccessful) identified a suggestive association signal on chromosome 28, indicating SORCS1 as a potential candidate gene associated with guide dog suitability (S1 Fig). Based on the preliminary GWAS results, we selected the top 12 SNPs in the canine SORCS1 gene with the strongest association signals for further analysis to evaluate whether the observed associations reflect true biological signals or false positives. To validate these associations, additional samples were collected, and association analyses between these 12 SNPs within SORCS1 and guide dog training outcomes were performed in a cohort of 64 dogs (32 successful and 32 unsuccessful). Significant differences in genotype frequencies between successful and unsuccessful guide dogs were observed for eight of the 12 SNPs—rs23432735, rs23419703, rs23433778, rs23402767, rs23402730, rs23402707, rs23418018, and rs23417942—located within the SORCS1 gene in predicted intron 3 (Genome Data Viewer; Canis lupus familiaris, CanFam3.1), when comparing major homozygotes versus all other genotypes or minor homozygotes versus all other genotypes (Table 2).
Pairwise linkage disequilibrium (LD) among these loci was evaluated by calculating D′ and r² values using genotype data from a cohort of 64 dogs.
Strong LD was observed among rs23432735, rs23419703, rs23433778, rs23402767, and rs23402730, with D′ values ranging from 0.867 to 1.000 and r² values from 0.678 to 1.000 (S3 Table). These SNPs are located in predicted intron 3 of the canine SORCS1 gene (NCBI Reference Sequences: NC_006610.3 and XM_005637877.2). Pairwise LD values (r²) are represented in an upper-triangular heat map shown in Fig 1.
Color intensity indicates the strength of LD, ranging from low (white) to high (red). The heat map depicts pairwise LD among 12 SNPs within the canine SORCS1 gene, evaluated in a cohort of 64 dogs comprising 32 successful and 32 unsuccessful guide dogs.
Based on the results obtained from the 64-dog cohort, we further expanded the sample size to confirm the association in a larger dataset. To confirm the association between SORCS1 genotype and guide dog training outcomes, we increased the sample size and genotyped dogs using a tagging SNP, rs23402730 (C > T), which showed one of the strongest statistical associations under a recessive model and was located within a region of high linkage disequilibrium with neighboring SNPs (Tables 2, S3 and Fig 1).
Genotype data for SORCS1 SNP rs23402730 were obtained from a cohort of 160 dogs (65 successful and 95 unsuccessful in training). Table 3 summarizes the association between SORCS1 SNP rs23402730 genotypes or alleles and guide dog training outcomes. The most significant difference was detected under a recessive genetic model (with the major allele considered recessive), when comparing CC + CT vs. TT genotypes (P = 0.0000474, OR = 4.11, 95% CI: 2.07–8.18).
Distribution of SORCS1 SNP rs23402730 in different dog breeds
We examined the allele frequency of the canine SORCS1 SNP rs23402730 in 220 dogs representing six breeds: Labrador Retrievers, Golden Retrievers, Chihuahuas, French Bulldogs, Toy Poodles, and Shiba Inus. Genotype and allele frequencies differed among breeds, with the frequency of the C allele ranging from 0.030 to 0.633 and 0.400 in total (S4 Table). These results indicate that SORCS1 SNP rs23402730 is broadly distributed across diverse dog breeds, with differences observed among breeds.
Association between SORCS1 and guide dog suitability after adjusting for covariates
Because allele frequencies of SORCS1 SNP rs23402730 differed considerably among breeds and several dogs in the training cohort were genetically related, we next evaluated the association between SORCS1 and behavioral suitability using a mixed-effects model that accounted for potential confounding by breed composition, sex, and family structure. The initial dataset included 160 dogs; however, one Standard Poodle was excluded because it was the only dog of that breed. The final sample used for the full analysis consisted of 159 dogs. In the full dataset of 159 dogs, the additive SORCS1 genotype was significantly associated with behavioral suitability after adjusting for sex, breed, and family effects (β = 2.20, SE = 0.63, P = 0.00045). The corresponding odds ratio was 8.98 (95% CI: 2.64–30.59), indicating that each additional allele increased the odds of successful guide dog qualification. Sex and breed did not show statistically significant contributions (Table 4). The variance component of the random effect (family) was substantial (SD = 1.62), indicating that clustering within family lines influenced behavioral outcomes and justified the use of a mixed model.
To further evaluate robustness in a genetically homogeneous background, we repeated the analysis using only the 114 Labrador retrievers. Even in the absence of breed effects, the additive genotype remained statistically significant (β = 1.55, SE = 0.71, P = 0.029), with an odds ratio of 4.71 (95% CI: 1.17–18.99). Sex was not associated with training outcome in either model (Table 5).
Taken together, these results indicated that the association between SORCS1 genotype and guide dog suitability remains significant after adjustment for sex, breed, and family relatedness, and is also observed within a more genetically uniform subgroup.
Preliminary screening for coding variation in SORCS1
As rs23402730 is located within predicted intron 3 of SORCS1, we conducted a preliminary screen of the predicted coding region to evaluate the potential contribution of exonic variants to functional differences. PCR amplification and sequencing of the coding regions were carried out in two dogs homozygous for the informative CC genotype. No functionally relevant coding variation was detected, suggesting that the observed association is unlikely to be explained by amino acid–altering mutations and may instead reflect regulatory effects.
Expression of canine SORCS1 gene in brain tissues
To determine whether the canine SORCS1 gene is expressed in the brain, RT-PCR was performed using cDNA derived from various canine tissues. SORCS1 cDNA was successfully amplified from several tissues, including the cerebral cortex, cerebellum, hypothalamus, hippocampus, and spinal cord (S2 Fig). These experiments were not designed to assess genotype-dependent expression differences; instead, they confirmed that the canine SORCS1 gene is transcribed in brain tissues, which supports the biological plausibility of behavioral associations.
Discussion
Behavioral suitability in guide dogs represents a complex, multifactorial phenotype. Although the present study analyzed 160 guide dogs, a modest sample size for genetic analyses of complex behavioral traits that may limit power to detect loci with small effects, the association between SORCS1 SNP rs23402730 and training outcome showed a comparatively large effect size and remained statistically significant. In addition, the association between rs23402730 and guide dog training outcomes remained statistically significant after adjustment for breed, sex, and relatedness. While guide dog suitability is not determined by any single gene, the present findings suggest that SORCS1 provides biologically and empirically meaningful information as part of the polygenic basis of behavioral suitability in working dogs.
The biological plausibility of SORCS1 as a behavioral candidate gene is supported by its high expression in the central nervous system and its established role in intracellular sorting and trafficking of synaptic receptors, including AMPA receptor complexes [25–27]. Previous studies have implicated SORCS1 and related Vps10p-domain receptors in synaptic plasticity and neurodegenerative disorders [28–32], and recent genomic analyses in foxes identified SORCS1 as a candidate gene associated with tame behavior [34]. Together, evidence from Alzheimer’s disease and fox domestication studies suggests that SORCS1 influences behavioral and neurological phenotypes primarily through haplotypes composed of noncoding variants that affect gene expression, rather than through protein-altering coding mutations. In this study, we suggested SORCS1 as a candidate gene associated with canine behavioral traits using SORCS1 intronic SNPs. However, its functional consequences remain uncertain in this study. Therefore, the current data do not establish a mechanistic explanation for the association between SORCS1 intronic SNPs and guide dog training outcomes. Future studies incorporating dense fine-mapping, haplotype analysis, allele-specific expression profiling, RNA-seq, and functional assays will be required to identify causal regulatory variants and to clarify how variation at the SORCS1 locus influences neural or behavioral phenotypes in dogs.
Limitations of the study
Clarifying the contribution of genetic diversity at the SORCS1 locus to guide dog suitability will require analyses in larger and more diverse populations, as well as assessment of unmeasured environmental factors, including housing conditions, handler effects, and training practices. Future prospective studies with standardized environmental and management data will be essential to disentangle genetic and environmental influences.
Conclusion
In this study, SORCS1 was identified as a candidate gene associated with behavioral traits relevant to guide dog suitability. Given the multifactorial nature of canine behavior, these findings contribute to understanding the genetic components of behavioral variation and may provide insight into the genetic architecture of working dog performance. More broadly, the results underscore the importance of continued research to identify causal variants and mechanisms underlying animal behavioral traits.
Supporting information
S1 Table. Samples from a genetic association study between canine SORCS1 SNPs and guide dog training outcomes.
https://doi.org/10.1371/journal.pone.0342346.s001
(XLSX)
S2 Table. List of primers used for canine SORCS1 gene analysis.
https://doi.org/10.1371/journal.pone.0342346.s002
(XLSX)
S3 Table. Pairwise linkage disequilibrium (LD) measures of |D′| and r² among 12 SNPs of the canine SORCS1 gene in a cohort of 64 dogs (32 successful and 32 unsuccessful guide dogs).
https://doi.org/10.1371/journal.pone.0342346.s003
(XLTX)
S4 Table. Distribution of the canine SORCS1 SNP rs23402730 among different dog breeds.
https://doi.org/10.1371/journal.pone.0342346.s004
(XLSX)
S1 Fig. Manhattan plot from a preliminary genome-wide association study (GWAS) assessing for guide dog training out come in a small canine cohort.
A suggestive association plot was detected on chromosome 28, indicating SORCS1 as a potential candidate gene associated with guide dog suitability.
https://doi.org/10.1371/journal.pone.0342346.s005
(PPTX)
S2 Fig. RT-PCR analysis of canine SORCS1 mRNA in various dog tissues.
Dog SORCS1 cDNA encompassing predicted exons 3–5 was amplified from different tissues. The integrity of RNA was confirmed using a primer pair for the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene.
https://doi.org/10.1371/journal.pone.0342346.s006
(PPTX)
Acknowledgments
All the authors thank to the staff in Kyushu Guide Dog Association and Prof. M. Bonkobara in Department of Veterinary Clinical Pathology, Nippon Veterinary and Life Science University (NVLU) for provide the samples. Also, author thank to Ms. N. Tada in Department of Basic Science, NVLU for their technical assistant.
References
- 1. Spady TC, Ostrander EA. Canine behavioral genetics: pointing out the phenotypes and herding up the genes. Am J Hum Genet. 2008;82(1):10–8. pmid:18179880
- 2. Rigterink A. Genetics of canine behavior: A review. World J Med Genet. 2014;4(3):46.
- 3. Mabunda RS, Makgahlela ML, Nephawe KA, Mtileni B. Evaluation of Genetic Diversity in Dog Breeds Using Pedigree and Molecular Analysis: A Review. Diversity. 2022;14(12):1054.
- 4. Mahmoodi M, Ayatollahi Mehrgardi A, Momen M, Serpell JA, Esmailizadeh A. Deciphering the genetic basis of behavioral traits in dogs: Observed-trait GWAS and latent-trait GWAS analysis reveal key genes and variants. Vet J. 2024;308:106251. pmid:39368730
- 5. Mohammadabadi M, Oleshko V, Oleshko O, Heiko L, Starostenko I, Kunovskii J, et al. Using Inter Simple Sequence Repeat Multi-Loci Markers for Studying Genetic Diversity in Guppy Fish. Turkish Journal of Fisheries and Aquatic Sciences. 2021;21(12):603–13.
- 6. Våge J, Wade C, Biagi T, Fatjó J, Amat M, Lindblad-Toh K, et al. Association of dopamine- and serotonin-related genes with canine aggression. Genes Brain Behav. 2010;9(4):372–8. pmid:20132314
- 7. Hejjas K, Vas J, Topal J, Szantai E, Ronai Z, Szekely A, et al. Association of polymorphisms in the dopamine D4 receptor gene and the activity‐impulsivity endophenotype in dogs. Animal Genetics. 2007;38(6):629–33.
- 8. Hejjas K, Kubinyi E, Ronai Z, Szekely A, Vas J, Miklósi A, et al. Molecular and behavioral analysis of the intron 2 repeat polymorphism in the canine dopamine D4 receptor gene. Genes Brain Behav. 2009;8(3):330–6. pmid:19382953
- 9. Wan M, Hejjas K, Ronai Z, Elek Z, Sasvari-Szekely M, Champagne FA, et al. DRD4 and TH gene polymorphisms are associated with activity, impulsivity and inattention in Siberian Husky dogs. Anim Genet. 2013;44(6):717–27. pmid:23713429
- 10. Kubinyi E, Vas J, Hejjas K, Ronai Z, Brúder I, Turcsán B, et al. Polymorphism in the tyrosine hydroxylase (TH) gene is associated with activity-impulsivity in German Shepherd Dogs. PLoS One. 2012;7(1):e30271. pmid:22272320
- 11. Lit L, Belanger JM, Boehm D, Lybarger N, Haverbeke A, Diederich C, et al. Characterization of a dopamine transporter polymorphism and behavior in Belgian Malinois. BMC Genet. 2013;14:45. pmid:23718893
- 12. Takeuchi Y, Kaneko F, Hashizume C, Masuda K, Ogata N, Maki T, et al. Association analysis between canine behavioural traits and genetic polymorphisms in the Shiba Inu breed. Anim Genet. 2009;40(5):616–22. pmid:19397510
- 13. Maejima M, Inoue-Murayama M, Tonosaki K, Matsuura N, Kato S, Saito Y, et al. Traits and genotypes may predict the successful training of drug detection dogs. Applied Animal Behaviour Science. 2007;107(3–4):287–98.
- 14. van den Berg L, Vos-Loohuis M, Schilder MBH, van Oost BA, Hazewinkel HAW, Wade CM, et al. Evaluation of the serotonergic genes htr1A, htr1B, htr2A, and slc6A4 in aggressive behavior of golden retriever dogs. Behav Genet. 2008;38(1):55–66. pmid:18066658
- 15. Takeuchi Y, Hashizume C, Arata S, Inoue-Murayama M, Maki T, Hart BL, et al. An approach to canine behavioural genetics employing guide dogs for the blind. Anim Genet. 2009;40(2):217–24. pmid:19170676
- 16. Luo D, Ma X, Bai J, Zhou Z, Wang F, Wang A, et al. Association between COMT SNP variation and timidity in Golden and Labrador Retrievers. Anim Genet. 2018;49(4):340–4. pmid:29932467
- 17. Konno A, Inoue-Murayama M, Hasegawa T. Androgen receptor gene polymorphisms are associated with aggression in Japanese Akita Inu. Biol Lett. 2011;7(5):658–60. pmid:21450727
- 18. Kis A, Bence M, Lakatos G, Pergel E, Turcsán B, Pluijmakers J, et al. Oxytocin receptor gene polymorphisms are associated with human directed social behavior in dogs (Canis familiaris). PLoS One. 2014;9(1):e83993. pmid:24454713
- 19. Persson ME, Trottier AJ, Bélteky J, Roth LSV, Jensen P. Intranasal oxytocin and a polymorphism in the oxytocin receptor gene are associated with human-directed social behavior in golden retriever dogs. Horm Behav. 2017;95:85–93. pmid:28765081
- 20. Cimarelli G, Virányi Z, Turcsán B, Rónai Z, Sasvári-Székely M, Bánlaki Z. Social Behavior of Pet Dogs Is Associated with Peripheral OXTR Methylation. Front Psychol. 2017;8:549. pmid:28443051
- 21. Konno A, Inoue-Murayama M, Yabuta S, Tonoike A, Nagasawa M, Mogi K, et al. Effect of Canine Oxytocin Receptor Gene Polymorphism on the Successful Training of Drug Detection Dogs. J Hered. 2018;109(5):566–72. pmid:29538681
- 22. Kovács K, Virányi Z, Kis A, Turcsán B, Hudecz Á, Marmota MT, et al. Dog-Owner Attachment Is Associated With Oxytocin Receptor Gene Polymorphisms in Both Parties. A Comparative Study on Austrian and Hungarian Border Collies. Front Psychol. 2018;9:435. pmid:29674985
- 23. Bai J, Chen S, Zhou Z, Zhao X, Song M, Guo Y, et al. Association between TH/MAOB gene single nucleotide polymorphisms and excitability in Labrador Retrievers. Anim Genet. 2023;54(4):570–5. pmid:37185969
- 24. Ding X, Hu J, Zhang H, Xu Y. Genetic Variants in the STMN1 Transcriptional Regulatory Region Affect Promoter Activity and Fear Behavior in English Springer Spaniels. PLoS One. 2016;11(7):e0158756. pmid:27390866
- 25. Hermey G, Riedel IB, Rezgaoui M, Westergaard UB, Schaller C, Hermans-Borgmeyer I. SorCS1, a member of the novel sorting receptor family, is localized in somata and dendrites of neurons throughout the murine brain. Neurosci Lett. 2001;313(1–2):83–7. pmid:11684345
- 26. Hermey G. The Vps10p-domain receptor family. Cell Mol Life Sci. 2009;66(16):2677–89. pmid:19434368
- 27. Savas JN, Ribeiro LF, Wierda KD, Wright R, DeNardo-Wilke LA, Rice HC, et al. The Sorting Receptor SorCS1 Regulates Trafficking of Neurexin and AMPA Receptors. Neuron. 2015;87(4):764–80. pmid:26291160
- 28. Liang X, Slifer M, Martin ER, Schnetz-Boutaud N, Bartlett J, Anderson B, et al. Genomic convergence to identify candidate genes for Alzheimer disease on chromosome 10. Hum Mutat. 2009;30(3):463–71. pmid:19241460
- 29. Lambert J-C, Amouyel P. Genetics of Alzheimer’s disease: new evidences for an old hypothesis? Curr Opin Genet Dev. 2011;21(3):295–301. pmid:21371880
- 30. Lane RF, Raines SM, Steele JW, Ehrlich ME, Lah JA, Small SA, et al. Diabetes-associated SorCS1 regulates Alzheimer’s amyloid-beta metabolism: evidence for involvement of SorL1 and the retromer complex. J Neurosci. 2010;30(39):13110–5. pmid:20881129
- 31. Reitz C, Tokuhiro S, Clark LN, Conrad C, Vonsattel J-P, Hazrati L-N, et al. SORCS1 alters amyloid precursor protein processing and variants may increase Alzheimer’s disease risk. Ann Neurol. 2011;69(1):47–64. pmid:21280075
- 32. Eggert S, Thomas C, Kins S, Hermey G. Trafficking in Alzheimer’s Disease: Modulation of APP Transport and Processing by the Transmembrane Proteins LRP1, SorLA, SorCS1c, Sortilin, and Calsyntenin. Mol Neurobiol. 2018;55(7):5809–29. pmid:29079999
- 33. Salasova A, Monti G, Andersen OM, Nykjaer A. Finding memo: versatile interactions of the VPS10p-Domain receptors in Alzheimer’s disease. Mol Neurodegener. 2022;17(1):74. pmid:36397124
- 34. Kukekova AV, Johnson JL, Xiang X, Feng S, Liu S, Rando HM, et al. Red fox genome assembly identifies genomic regions associated with tame and aggressive behaviours. Nat Ecol Evol. 2018;2(9):1479–91. pmid:30082739
- 35.
Cabinet Office. White paper on persons with disabilities 2025 (in Japanese). Japan government; 2025.
- 36. Serpell JA, Hsu Y, Hiby EF. Development and validation of a method for evaluating guide dog behavior. Appl Anim Behav Sci. 2009;121:201–13.
- 37. Bray EE, Sammel MD, Cheney DL, Serpell JA, Seyfarth RM. Effects of maternal investment, temperament, and cognition on guide dog success. Proc Natl Acad Sci USA. 2017;114(34):9128–33.
- 38. Goddard ME, Beilharz RG. Genetic and environmental factors affecting the suitability of dogs as Guide Dogs for the Blind. Theor Appl Genet. 1982;62(2):97–102. pmid:24270555
- 39. Batt LS, Batt MS, Baguley JA, McGreevy PD. Factors associated with success in guide dog training. Journal of Veterinary Behavior. 2008;3(4):143–51.
- 40. Arata S, Momozawa Y, Takeuchi Y, Mori Y. Behavioral characteristics predict guide dog qualification. Sci Rep. 2015;5:1–10.
- 41. Wilsson E, Sundgren P-E. The use of a behaviour test for the selection of dogs for service and breeding, I: Method of testing and evaluating test results in the adult dog, demands on different kinds of service dogs, sex and breed differences. Applied Animal Behaviour Science. 1997;53(4):279–95.
- 42.
Japan Guide Dog Association Council. Guide dog instructor training textbook. In: Japan guide Dog Facilities Association. Japan; 2003.
- 43. Park S-J, Huh J-W, Kim Y-H, Lee S-R, Kim S-H, Kim S-U, et al. Selection of internal reference genes for normalization of quantitative reverse transcription polymerase chain reaction (qRT-PCR) analysis in the canine brain and other organs. Mol Biotechnol. 2013;54(1):47–57. pmid:22531949