Dodo6/Topic_Modelling_using_LDA

1Article2 3Genome-wide Analyses Identifies Known and New4Markers Responsible of Chicken Plumage Color5Salvatore Mastrangelo 1, Filippo Cendron 2,*, Gianluca Sottile 3, Giovanni Niero 2,6Baldassare Portolano 1, Filippo Biscarini 4 and Martino Cassandro 27Dipartimento Scienze Agrarie, Alimentari e Forestali, University of Palermo, 90128 Palermo, Italy;8salvatore.mastrangelo@unipa.it (S.M.); baldassare.portolano@unipa.it (B.P.)92 Dipartimento di Agronomia Animali Alimenti Risorse naturali e Ambiente, University of Padova,1035020 Legnaro, Italy; filippo.cendron@phd.unipd.it (F.C.); g.niero@unipd.it (G.N.);11martino.cassandro@unipd.it (M.C.)123 Dipartimento Scienze Economiche, Aziendali e Statistiche, University of Palermo, 90128 Palermo, Italy;13gianluca.sottile@unipa.it144 CNR-IBBA, 20133 Milano, Italy; filippo.biscarini@gmail.com15* Correspondence: filippo.cendron@phd.unipd.it; Tel.: +39-349762736816117 18Received: 5 February 2020; Accepted: 14 March 2020; Published: 15 March 202019 20Simple Summary: In order to assess sources of variation related to Polverara breed plumage color21(black vs. white), we carried out genome-wide analyses to identify the genomic regions involved in22this trait. The present work has revealed new candidate genes involved in the phenotypic variability23in local chicken populations. These results also contribute insights into the genetic basis for plumage24color in poultry, and confirm the great complexity of the mechanisms that control this trait.25Abstract: Through the development of the high-throughput genotyping arrays, molecular markers26and genes related to phenotypic traits have been identified in livestock species. In poultry, plumage27color is an important qualitative trait that can be used as phenotypic marker for breed identification.28In order to assess sources of genetic variation related to the Polverara chicken breed plumage colour29(black vs. white), we carried out a genome-wide association study (GWAS) and a genome-wide30fixation index (FST) scan to uncover the genomic regions involved. A total of 37 animals (17 white31and 20 black) were genotyped with the Affymetrix 600 K Chicken single nucleotide polymorphism32(SNP) Array. The combination of results from GWAS and FST revealed a total of 40 significant33markers distributed on GGA 01, 03, 08, 12 and 21, and located within or near known genes. In34addition to the well-known TYR, other candidate genes have been identified in this study, such as35GRM5, RAB38 and NOTCH2. All these genes could explain the difference between the two Polverara36breeds. Therefore, this study provides the basis for further investigation of the genetic mechanisms37involved in plumage color in chicken.38Keywords: local chicken populations; genome-wide analyses; SNP; plumage color; candidate genes39 401. Introduction41Over the last century, erosion of livestock genetic resources has been observed as a result of the42massive replacement of low-productivity local breeds with highly productive ones. These local43breeds are nonetheless an important reservoir of genetic diversity, each with specific characteristics.44Local animal genetic resources might indeed be characterized by specific heritable phenotypes45potentially relevant for current or future use in breeding programs [1]. Several studies showed that46 47Animals 2020, 10, 493; doi:10.3390/ani1003049348 49www.mdpi.com/journal/animals50 51Animals 2020, 10, 49352 532 of 954 55local populations can be useful for the investigation of the genetic factors underlying those unique56phenotypes related to their diversity [2–4].57In Italy, there are numerous known local chicken breeds whose overall conservation status is58nevertheless critical; with the abandoning of farming in marginal areas and the advent of industrialscale chicken breeding, highly specialized chicken lines have replaced the less competitive local59breeds [5]. The interest in the conservation of Italian local chicken breeds emerged from an in situ60marker-assisted conservation scheme, that involved seven breeds reared in region Veneto:61Ermellinata di Rovigo, Pepoi, Robusta Lionata, Robusta Maculata, Millefiori di Lonigo, Padovana62and Polverara. The latter is an ancient dual-purpose chicken breed, named after a small town south63of Padua. The early history of the Polverara breed is unclear, but it is believed to be the result of a64cross between Padovana and other local Veneto chicken populations [6]. The Polverara is a mediumsized chicken with a feathery crest, that erects over the head without covering the eyes. Two different65monochrome plumage colors are officially recognized for the Polverara breed, black and white,66resulting in two populations: Polverara White (PW) and Polverara Black (PB). Additional Polverara67color-varieties may result from crossbreeding between the breed with other local fowls, but they are68not standardized. PW and PB are reared separately, and cross‐breeding is not commonly practiced,69or at least not recorded. Evidence from previous studies shows close genetic relationships between70the two Polverara populations [5,7].71As a consequence of their features (phenotypic differentiation and common genetic72background), these two populations provide an interesting model to study the genomic regions73underpinning their phenotypic diversity, in particular the plumage color.74Alongside the advance of high-throughput genotyping arrays, molecular markers and genes75associated to phenotypic traits or diseases in chickens have been identified through genome-wide76approaches [8–10]. In this study, we carried out a genome-wide association study (GWAS) and a77genome-wide fixation index (FST) scan to identify genomic regions that may explain the phenotypic78differences observed between PW and PB.792. Materials and Methods802.1. DNA Samples, Genotyping and Quality Control81The collection of blood samples was conducted as part of routine health screening by qualified82veterinarians following guidelines established by Institutional Animal Care and Use Committee83(IACUC).84Blood samples were collected from ulnar veins from 37 unrelated animals belonging to the85Polverara White (PW) (n = 17) and Polverara Black (PB) (n = 20) chicken breeds (Figure 1). The animals86were randomly selected from three different conservation centers located in different areas of Veneto.87DNA samples were genotyped using Affymetrix Axiom 600 K Chicken Genotyping Array containing88580,954 single nucleotide polymorphisms (SNPs). The Gallus_gallus-5.0 chicken genome assembly89was used in this study as a reference. Only markers located on chromosomes 1 to 28 were used.90Quality control procedures were performed for the genotype data using PLINK 1.9 [11]. The91following filtering parameters were adopted: (i) SNPs with call rate <95%, (ii) minor allele frequency92<5% and (iii) animals with more than 10% of missing genotypes were removed.93 94Animals 2020, 10, 49395 963 of 997 98Figure 1. Specimens of Polverara White (PW) and Polverara Black (PB) chickens.99 1002.2. Genome-wide Analyses101We performed a genome-wide association study (GWAS) using the univariate case-control102model (PW vs. PB) implemented in the snpassoc R package [12], specifically the log-additive genetic103model. We used Bonferroni correction to determine the genome-wide significance threshold defined104as 0.0001/N (N being the number of tested SNPs).105The FST case-control analysis was performed using the –fst functionality in PLINK 1.9 [11], by106comparing single markers between the PW and PB. Relevant FST differences were defined considering107the SNPs falling in and above the 99.98th percentile distribution [3,4].108A Manhattan plot of the results was generated using the R package qqman [13]. p- and FST values109of each SNP were plotted as a function of its position along each autosomal chromosome. The110overlapping genomic regions identified by both approaches were further explored to identify linked111candidate112genes113using114the115Genome116Data117Viewer118(https://www.ncbi.nlm.nih.gov/genome/gdv/browser/genome/?id=GCF_000002315.4) developed by119NCBI. To investigate the biological functions and the phenotypes that are known to be regulated by120each annotated gene, we conducted a comprehensive literature search, including information from121other species. Pair-wise Linkage Disequilibrium (LD) was estimated as the genotype correlation122coefficient (r2) [14]. For all pairs of autosomal SNPs, r2 measures were obtained using the–r2–ldwindow 99999–ld-window-r2 0 command in PLINK v1.9 [11]. LD values were grouped into bins123based on the base-pair distance between SNPs from the physical map. The average per-bin LD as a124function of the base-pair distance was then used to estimate LD decay.1253. Results126After quality control (see above), the final number of SNPs retained for the analysis was 283,893127and no animal was discarded due to poor quality genotyping.128The GWAS analysis revealed a total of 80 highly significant Bonferroni corrected SNPs (p <1290.0001 (−log10 (p) = 9.45) located on eight autosomes (Table S1). The corresponding Manhattan plot130is reported in Figure 2a. The chicken chromosome (GGA) 01 showed the largest number of significant131markers (55), and except for one marker, all the SNPs on this chromosome were located inside a 3,57132Mb region (184,995,531–188,565,711 bp) (Table S1). Moreover, these markers on GGA01 are plotted133in two single points on the Manhattan plot because they are adjacent to each other and had the same134p-value (Figure 2a).135To further support results from GWAS, a genome-wide FST case-control analysis was also136performed. The analysis showed a total of 66 SNPs above the selected threshold (FST = 0.74), located137on six different autosomes (GGA 01, 03,08,12,14 and 21) (Table S2; Figure 2b). In agreement to the138results provided by GWAS, the highest number of significant markers are mapped on GGA01 (52).139Combining the results from GWAS and FST, we identified a total of 40 significant markers140distributed on GGA 01, 03, 08, 12 and 21 (Table 1).141Levels of pairwise LD decreased with increasing genomic distance between SNPs (Figure 3). The142Polverara breed showed moderate LD decay, with the average r2 falling below 0.20 after 50 Kb.143 144Animals 2020, 10, 493145 1464 of 9147 148Figure 2. (a) Manhattan plot of the p-values in the genome-wide association study (GWAS). The149horizontal lines represent the Bonferroni-corrected genome-wide significance (red; p < 0.0001); (b)150Manhattan plot of the genome-wide fixation index (FST). The horizontal line represents the genomewide significance single nucleotide polymorphisms (SNP) above the 99.98th percentile distribution)151(FST = 0.74). Significant SNPs are highlighted in green.152 153Figure 3. Linkage Disequilibrium decay (measured as r2) as a function of inter-marker distance (Kbp)154in the Polverara breed.155 156Animals 2020, 10, 493157 1585 of 9159 160Several SNPs were adjacent or near to each other. We searched candidate genes within 250 kblong regions (125 kb upstream and 125 kb downstream) around peak SNPs, which corresponded to161median r2 ≥ 0.16. A total of 17 known genes were identified (Table 1).162Table 1. Overlapping significant markers identified by GWAS and FST and associated genes.163GGA164116511661167116811691170117111721173117411751176117711781179118011811182118311841185118611871188118911901191319231938194819581968197819881991220012201122022120321204 205SNP206AX-75371751207AX-75373909208AX-75374539209AX-75375587210AX-75376255211AX-75376262212AX-75378645213AX-75378836214AX-75378888215AX-75379333216AX-75379334217AX-75379450218AX-77278759219AX-75379693220AX-75379724221AX-75379753222AX-75379761223AX-75379775224AX-75379792225AX-75379800226AX-75379813227AX-75380172228AX-75380766229AX-75380808230AX-80852333231AX-75380927232AX-75380931233AX-76506116234AX-76506117235AX-77109355236AX-77109358237AX-77109696238AX-77109700239AX-77109855240AX-77109898241AX-75680106242AX-75680164243AX-75680170244AX-76239008245AX-76239099246 247Position (bp)248184995531249185836576250186058014251186464423252186722445253186735600254187660456255187723578256187743605257187911192258187911433259187960805260188025840261188066880262188079273263188089989264188093458265188096972266188102761267188106002268188112765269188238879270188476552271188490865272188493037273188538625274188540546275559295332765593017827740129062784014014279416438428041679842814230320282424945028310597665284106274732851062957928626402992872657895288 289p-Value2905.45e-112915.45e-112925.45e-112935.45e-112945.45e-112955.45e-112969.01e-132979.01e-132989.01e-132999.01e-133009.01e-133019.01e-133029.01e-133039.01e-133049.01e-133059.01e-133069.01e-133079.01e-133089.01e-133099.01e-133109.01e-133119.01e-133129.01e-133139.01e-133149.01e-133159.01e-133169.01e-133175.45e-113185.45e-113192e-103202e-103212e-103222e-103232e-103242e-103252e-103262e-103272e-103282e-103292e-10330 331FST3320.7453330.7733340.8583350.8583360.8863370.8863380.8093390.8093400.8093410.8093420.8093430.8093440.8093450.8093460.8093470.8093480.8093490.8093500.8093510.8093520.8093530.8093540.8093550.8093560.8093570.8093580.8093590.7453600.7453610.7573620.7573630.7573640.7573650.7573660.7573670.7553680.7553690.7553700.7573710.757372 373Nearest Gene374Name375Distance (kb)376MAML23772.01378LOC1070523493792.90380CCDC6738129.43382FAT3383 384Within385 386NAALAD2387FOLH1388FOLH1389NOX4390NOX4391TYR392 393Within3942.9039522.923968.023978.26398Within399 400GRM5401 402Within403 404RAB38405RAB38406RAB38407TMEM135408TMEM135409HBS1L410HBS1L411CRIP1412CRIP1413 41488.89415103.21416105.3841761.5141859.594199.404209.404217.014228.12423 424SEC22B425 426Within427 428NOTCH2429 430Within431 432KLF15433KLF15434KLF15435C21H1ORF159436C21H1ORF159437 43837.984398.174406.074410.534421.85443 444Note: Gallus gallus chromosome number, GGA; single nucleotide polymorphism, SNP.445 4464. Discussion447Potentially, there is much unrecognized beneficial genetic variation in local autochthonous448animal breeds and populations [15]. As visual characteristics of animals, pigmentation traits are often449used for breed identification, and represent an important phenotype of interest for breeding and450research [16]. Several genome-wide studies for coat color have been conducted in livestock species451including cattle [16–18], sheep [2,19], goat [20,21]. In this work, genome-wide analyses have been452 453Animals 2020, 10, 493454 4556 of 9456 457performed in the Polverara chicken breed, PW and PB subpopulations. Considering its phenotypic458variability (black vs. white), this local breed has been used as a model for investigating the genetic459bases of plumage color.460Based on the greater power of analysis in limiting the number of false positive signals when461more than one methodology is adopted in parallel, two different approaches (GWAS and FST) have462been used in this study [3,4]. From results, the major overlap in genomic regions associated to the463phenotypic differences was found on GGA01 (Table 1). The most striking result refers to a relatively464narrow 0.88 Mb interval (187,660,456–188,540,546 bp). This region showed strong divergence465between PW and PB. One of the most significant markers (SNP AX-75379450) was located within the466TYR gene, while a total of 10 significant SNPs were concentrated in a very small interval of 0.27 Mb467within the GRM5 gene. TYR codes for a key enzyme in melanin biosynthesis and it has been accepted468as a major gene involved in plumage color in chickens [22–24]. A previous study reported that TYR469showed the greatest level of differential expression in the skin of black versus white chickens [24]. In470humans [25–27] and mice [28], several genome-wide studies have also shown signals of association471for skin or coat color in the genomic regions encompassing the GRM5 and TYR genes. On GGA1,472there were three other significant markers close to the RAB38 gene whose products is a Ras-related473protein. Ras-related proteins are critical regulators of cellular membrane trafficking [29] and are474involved in a variety of processes, including skin pigmentation [30]. It has been reported that the475mouse RAB38 gene acts in a functionally redundant way in regulating skin melanocyte pigmentation476and controls the post-Golgi trafficking of tyrosinase (TYR) and tyrosinase-related protein 1 (TYRP1)477[31]. Moreover, a GWAS for chicken plumage pigmentation reported a gene belonging to the RAS478family, RAS4A, located in the region of a significantly associated SNP [10]. On GGA08, two SNPs479(AX-77109855 and AX-77109898) were both located within the NOTCH2 gene. A recent study [32]480reported that Notch signaling is involved in the regulation of melanocyte development during481adulthood, and NOTCH2 contributes to the regulation of melanocyte homeostasis. Furthermore,482NOTCH2 cooperates with c-kit signaling during embryogenesis, and they cooperate to regulate483melanocyte homeostasis after birth [32]. Therefore, in addition to the well-known TYR gene, it can be484hypothesized that variants of the GRM5, RAB38 and NOTCH2 genes could be related to plumage485color in chicken. A previous GWAS for plumage color [33], using a low-density array, revealed a486significant association with SNPs mapped on the AKT3, KRT7, PAP2 and DDX6 genes. Yang et al.,487[10], in a GWAS using black and no-black chickens, showed a strong association with SNPs within488SHH and NUAK genes, while Johansson and Nelson [34] reported that the EDN3 gene is associated489with dark pigmentation in two local chickens breeds. The authors did not observe any association490with the candidate genes here reported. A possible reason for the lack of correspondence among491studies may be the different breeds used in the comparison (and their plumage color), the array492density and the statistical approaches. In this study, we have reported as candidate loci for chicken493plumage color the genomic regions obtained combining the results from two different approaches494applied to PW and PB. Despite the phenotypic differences, the two populations share a common495genetic background [5–7]. This leads to minimize the confounding effects due to genetic divergence496and population structure [15,20]. Moreover, some candidate genes identified here, such TYR, are497consistent with results reported from previous studies on chicken plumage color. All of the above is498likely to have helped us circumvent potential biases linked to false positive signals: the identified499genes should therefore be considered rather robust results, which can contribute to explain the500genetic contribution to phenotypic differences between PW and PB.501The most obvious phenotypic difference between PW and PB is the plumage color, and a number502of genes involved in the determination of this phenotype have been detected in this study. It should503also be pointed out that other known genes have been identified by combining GWAS and FST.504Significant markers on GGA01 were close to candidate genes involved in feed conversion ratio505(NOX4) [35] and feed efficiency (TMEM135) [36] in chickens. On GGA12, the analyses revealed three506markers close to KLF15, a gene associated with chicken growth and carcass traits [37]. It is likely that507the two populations differ for additional less obvious phenotypes, such as reproductive performance508or feed efficiency.509 510Animals 2020, 10, 493511 5127 of 9513 5145. Conclusions515In poultry, plumage color is an important qualitative trait that can serve as marker useful for516breed identification. Although the chicken genome is well studied, not all the genes affecting517plumage color are described. Based on previous studies in other species, the present work has518revealed new potential candidate genes involved in the phenotypic variability of color in local519chicken populations. These results contribute insights into the genetic basis for plumage color in520poultry, and confirm the great complexity of the mechanisms that control this trait. Additional521research will be necessary to refine the presented results and further investigate the molecular522mechanism underpinning plumage color.523Supplementary Materials: The following are available online at www.mdpi.com/2076-2615/10/3/493/s1,524Table S1 List of significant single nucleotide polymorphisms (SNPs) obtained in the genome-wide association525study (GWAS) (p < 0.0001 (−log10 (p) = 9.45). Table S2 List of significant single nucleotide polymorphisms (SNPs)526obtained in the genome-wide fixation index (FST ≥ 0.74).527Author Contributions: Conceptualization, S.M., and M.C.; methodology, S.M., G.S., F.B., and G.N.; formal528analysis, S.M., F.B., and G.S.; investigation, S.M., B.P. and M.C; resources, M.C.; data curation, S.M., F.C., and529M.C.; writing—original draft preparation, S.M.; writing—review and editing, S.M., F.C., G.S., G.N., B.P., F.B.,530and M.C.; supervision, B.P., and M.C.; funding acquisition, M.C. F.B. is currently seconded at the ERCEA531(European Research Council Executive Agency), Brussels, Belgium. The views expressed here are purely those532of the writer and may not in any circumstances be regarded as stating an official position of the European533Commission. All authors have read and agreed to the published version of the manuscript.534Funding: This work was supported by the project: “Protection of biodiversity of Italian poultry breeds" TuBaVi535- 2014 – 2020, PSRN - Support for the conservation, use and sustainable development of genetic resources in536agriculture, sub-measure 10.2. https://www.pollitaliani.it/en/project/.537Acknowledgments: We acknowledge the conservation centers: I.I.S. “Duca degli Abruzzi” di Padova, I.S.I.S.S.538“D. Sartor” di Castelfranco Veneto (Treviso), I.I.S. “A. Della Lucia” di Feltre (Belluno), Azienda Sperimentale539“Sasse Rami” di Ceregnano (Rovigo) for proving blood samples of poultry breeds.540Conflicts of Interest: The authors declare no conflict of interest.541 542References5431.5442.545 5463.547 5484.549 5505.5516.5527.553 5548.555 556Leroy, G.; Besbes, B.; Boettcher, P.; Hoffmann, I.; Capitan, A.; Baumung, R. Rare phenotypes in domestic557animals: Unique resources for multiple applications. Anim. Genet. 2016, 47, 141–153.558Kijas, J.W.; Serrano, M.; McCulloch, R.; Li, Y.; Salces Ortiz, J.; Calvo, J.H.; Pérez-Guzmán, M.D. The559International Sheep Genomics Consortium. Genome wide association for a dominant pigmentation gene560in sheep. J. Anim. Breed. Genet. 2013, 130, 468–475.561Schiavo, G.; Bertolini, F.; Utzeri, V.J.; Ribani, A.; Geraci, C.; Santoro, L.; Ovilo, C.; Fernandez, A.I.; Gallo,562M.; Fontanesi, L. Taking advantage from phenotype variability in a local animal genetic resource:563Identification of genomic regions associated with the hairless phenotype in Casertana pigs. Anim. Genet.5642018, 49, 321–325.565Mastrangelo, S.; Sottile, G.; Sutera, A.M.; Di Gerlando, R.; Tolone, M.; Moscarelli, A.; Sardina, M.T.;566Portolano, B. Genome-wide association study reveals the locus responsible for microtia in Valle del Belice567sheep breed. Anim. Genet. 2018, 49, 636–640.568Zanetti, E.; De Marchi, M.; Dalvit, C.; Cassandro, M. Genetic characterization of local Italian breeds of569chickens undergoing in situ conservation. Poult. Sci. 2010, 89, 420–427.570De Marchi, M.; Dalvit, C.; Targhetta, C.; Cassandro, M. Assessing genetic variability in two ancient chicken571breeds of Padova area. Ital. J. Anim. Sci. 2005, 4 (Suppl. 3), 151–153.572Viale, E.; Zanetti, E.; Özdemir, D.; Broccanello, C.; Dalmasso, A.; De Marchi, M.; Cassandro, M.573Development and validation of a novel SNP panel for the genetic characterization of Italian chicken breeds574by next-generation sequencing discovery and array genotyping. Poult. Sci. 2017, 96, 3858–3866.575Bai, H.; Sun, Y.; Liu, N.; Xue, F.; Li, Y.; Xu, S.; Ye, J.; Zhang, L.; Chen, Y.; Chen, J. Single SNP-and pathwaybased genome-wide association studies for beak deformity in chickens using high-density 600K SNP576arrays. BMC Genom. 2018, 19, 501.577 578Animals 2020, 10, 493579 5809.58110.58211.58312.58413.58514.58615.587 58816.58917.59018.591 59219.593 59420.595 59621.597 59822.599 60023.601 60224.60325.604 60526.606 60727.608 60928.610 61129.612 6138 of 9614 615Wang, Q.; Pi, J.; Shen, J.; Pan, A.; Qu, L. Genome-wide association study confirms that the chromosome Z616harbours a region responsible for rumplessness in Hongshan chickens. Anim. Genet. 2018, 49, 326–328.617Yang, L.; Du, X.; Wei, S.; Gu, L.; Li, N.; Gong, Y.; Li, S. Genome-wide association analysis identifies potential618regulatory genes for eumelanin pigmentation in chicken plumage. Anim. Genet. 2017, 48, 611–614.619Chang, C.C.; Chow, C.C.; Tellier, L.C.; Vattikuti, S.; Purcell, S.M.; Lee, J.J. Second-generation PLINK: Rising620to the challenge of larger and richer datasets. Gigascience 2015, 4, 7.621Gonzalez, J.R.; Armengol, L.; Sole, X.; Guino, E.; Mercader, J.M.; Estivill, X.; Moreno, V. SNPASSOC: An R622package to perform whole genome association studies. Bioinformatics 2007, 23, 654–655.623Turner, S.D. qqman: An R package for visualizing GWAS results using QQ and manhattan plots. Biorxiv6242014, 2, 005165.625VanLiere, J.M.; Rosenberg, N.A. Mathematical properties of the r2 measure of linkage disequilibrium. Theor.626Popul. Biol. 2008, 74, 130–137.627Mastrangelo, S.; Ben Jemaa, S.; Sottile, G.; Casu, S.; Portolano, B.; Ciani, E.; Pilla, F. Combined approaches628to identify genomic regions involved in phenotypic differentiation between low divergent breeds:629Application in Sardinian sheep populations. J. Anim. Breed. Genet. 2019, 136, 526–534.630Fan, Y.; Wang, P.; Fu, W.; Dong, T.; Qi, C.; Liu, L.; Guo, G.; Li, C.; Ciu, X.; Zhang, S.; et al. Genome-wide631association study for pigmentation traits in Chinese Holstein population. Anim. Genet. 2014, 45, 740–744.632Edea, Z.; Dadi, H.; Dessie, T.; Kim, I.H.; Kim, K.S. Association of MITF loci with coat color spotting patterns633in Ethiopian cattle. Genes Genom. 2017, 39, 285–293.634Mastrangelo, S.; Sottile, G.; Sardina, M.T.; Sutera, A.M.; Tolone, M.; Di Gerlando, R.; Portolano, B. A635combined genome-wide approach identifies a new potential candidate marker associated with the coat636color sidedness in cattle. Livest. Sci. 2019, 225, 91–95.637Muniz, M.M.M.; Caetano, A.R.; McManus, C.; Cavalcanti, L.C.G.; Façanha, D.A.E.; Leite, J.H.G.M.; Facò,638O.; Paiva, S.R. Application of genomic data to assist a community- based breeding program: A preliminary639study of coat color genetics in Morada Nova sheep. Livest. Sci. 2016, 190, 89–93.640Kumar, C.; Song, S.; Dewani, P.; Kumar, M.; Parkash, O.; Ma, Y.; Jiang, L. Population structure, genetic641diversity and selection signatures within seven indigenous Pakistani goat populations. Anim. Genet. 2018,64249, 592–604.643Nazari-Ghadikolaei, A.; Mehrabani-Yeganeh, H.; Miarei-Aashtiani, S.R.; Staiger, E.A.; Rashidi, A.; Huson,644H.J. Genome-wide association studies identify candidate genes for coat color and Mohair Traits in the645Iranian Markhoz Goat. Front. Genet. 2018, 9, 105.646Chang, C.M.; Coville, J.L.; Coquerelle, G.; Gourichon, D.; Oulmouden, A.; Tixier-Boichard, M. Complete647association between a retroviral insertion in the tyrosinase gene and the recessive white mutation in648chickens. BMC Genom. 2006, 7, 19.649Liu, W.B.; Chen, S.R.; Zheng, J.X.; Qu, L.J.; Xu, G.Y.; Yang, N. Developmental phenotypic-genotypic650associations of tyrosinase and melanocortin 1 receptor genes with changing profiles in chicken plumage651pigmentation. Poult. Sci. 2010, 89, 1110–1114.652Zhang, J.; Liu, F.; Cao, J.; Liu, X. Skin transcriptome profiles associated with skin color in chickens. PLoS653ONE 2015, 10, e0127301.654Beleza, S.; Johnson, N.A.; Candille, S.I.; Absher, D.M.; Coram, M.A.; Lopes, J.; Campos, J.; Araùjo, I.I.;655Anderson, T.M.; Vilhjàlmsson, B.J.; et al. Genetic architecture of skin and eye color in an African-European656admixed population. PLoS Genet. 2013, 9, e1003372.657Lloyd-Jones, L.R.; Robinson, M.R.; Moser, G.; Zeng, J.; Beleza, S.; Barsh, G.S.; Tang, H.; Visscher, P.M.658Inference on the genetic basis of eye and skin color in an admixed population via Bayesian linear mixed659models. Genetics 2017, 206, 1113–1126.660Lona-Durazo, F.; Hernandez-Pacheco, N.; Fan, S.; Zhang, T.; Choi, J.; Kovacs, M.A.; Loftus, S.K.; Le, P.;661Edwards, M.; Fortes-Lima, C.A.; et al. Meta-analysis of GWA studies provides new insights on the genetic662architecture of skin pigmentation in recently admixed populations. BMC Genet. 2019, 20, 59.663Miao, Y.; Soudy, F.; Xu, Z.; Liao, M.; Zhao, S.; Li, X. Candidate Gene Identification of Feed Efficiency and664Coat Color Traits in a C57BL/6J× Kunming F2 Mice Population Using Genome-Wide Association Study.665BioMed Res. Int. 2017, 2017, doi:10.1155/2017/7132941.666Pfeffer, S.R. Rab GTPases: Specifying and deciphering organelle identity and function. Trends Cell Biol. 2001,66711, 487–491.668 669Animals 2020, 10, 493670 67130.67231.673 67432.675 67633.67734.67835.679 68036.681 68237.683 6849 of 9685 686Wasmeier, C.; Romao, M.; Plowright, L.; Bennett, D.C.; Raposo, G.; Seabra, M.C. Rab38 and Rab32 control687post-Golgi trafficking of melanogenic enzymes. J. Cell Biol. 2006, 175, 271–281.688Brooks, B.P.; Larson, D.M.; Chan, C.C.; Kjellstrom, S.; Smith, R.S.; Crawford, M.A.; Lamoreux, L.; Huizing,689M.; Hess, R.; Jiao, X. Analysis of ocular hypopigmentation in Rab38cht/cht mice. Investig. Ophthalmol. Vis.690Sci. 2007, 48, 3905–3913.691Kumano, K.; Masuda, S.; Sata, M.; Saito, T.; Lee, S.Y.; Sakata-Yanagimoto, M.; Tomita, T.; Iwatsubo, T.;692Natsugari, H.; Kurokawa, M.; et al. Both Notch1 and Notch2 contribute to the regulation of melanocyte693homeostasis. Pigment. Cell Melanoma Res. 2008, 21, 70–78.694Park, M.N.; Choi, J.A.; Lee, K.T.; Lee, H.J.; Choi, B.H.; Kim, H.; Kim, T.H.; Cho, S.; Lee, T. Genome-wide695association study of chicken plumage pigmentation. Asian-Aust. J. Anim. Sci. 2013, 26, 1523.696Johansson, A.M.; Nelson, R.M. Characterization of genetic diversity and gene mapping in two Swedish697local chicken breeds. Front. Genet. 2015, 6, 44.698Shah, T.M.; Patel, N.V.; Patel, A.B.; Upadhyay, M.R.; Mohapatra, A.; Singh, K.M.; Deshpande, S.D.; Joshi,699C.G. A genome-wide approach to screen for genetic variants in broilers (Gallus gallus) with divergent feed700conversion ratio. Mol. Genet. Genom. 2016, 291, 1715–1725.701Lee, J.; Karnuah, A.B.; Rekaya, R.; Anthony, N.B.; Aggrey, S.E. Transcriptomic analysis to elucidate the702molecular mechanisms that underlie feed efficiency in meat-type chickens. Mol. Genet. Genom. 2015, 290,7031673–1682.704Lyu, S.J.; Tian, Y.D.; Wang, S.H.; Han, R.L.; Mei, X.X.; Kang, X.T. A novel 2-bp indel within Krüppel-like705factor 15 gene (KLF15) and its associations with chicken growth and carcass traits. Brit. Poult. Sci. 2014, 55,706427–434.707© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access708article distributed under the terms and conditions of the Creative Commons Attribution709(CC BY) license (http://creativecommons.org/licenses/by/4.0/).710 711