Quantitative trait loci for morphometric and mineral composition traits of the tibia bone in a broiler × layer cross.

Many economic losses occur in the poultry industry due to leg fragility. Knowing the genomic regions that influence traits associated with the growth and composition of the leg's bone can help to improve the selection process leading to increased leg resistance to fracture. The present study aimed to map quantitative trait loci (QTL) for mineral composition and morphometric traits of the tibia in 478 animals from an F2 broiler × layer cross. The measurement of weight, length and width of Tibia was carried out at 42 days of age. Ash, dry matter, levels of calcium (Ca), phosphorus (P), magnesium (Mg), Zinc (Zn) and Calcium:Phosphorus (Ca:P) ratio were also recorded. The population was genotyped for 128 microsatellite markers and one single nucleotide polymorphism, covering 2630 cM of the chicken genome. A likelihood ratio test was performed to find QTLs. Additive and dominance effects of the QTLs were included in the model. In the chromosomes 2 (GGA2), 6 (GGA6), 8 (GGA8), 24 (GGA24) and 26 (GGA26) some suggestive QTLs (P<0.00276) were mapped for tibia weight (GGA2 and GGA26), ash percentage (GGA2 and GGA6), dry matter percentage (GGA2), Ca (GGA8 and GGA24) and Ca:P ratio (GGA8), many of which are close to genes already identified as good candidates for those traits. The suggestive QTL on GGA2 has a pleiotropic effect on ash percentage, dry matter and bone weight, whereas in the GGA8 there seems to be two QTLs, one for Ca and another for Ca:P ratio. Thus, this study identified at least five genomic regions, in different chromosomes, that can be targeted for further research to identify potential mutations influencing the development and composition of leg bones in Gallus gallus.

Whole transcriptome analysis of the pectoralis major muscle reveals molecular mechanisms involved with white striping in broiler chickens.

White striping (WS) is one of the most common myopathies identified in broiler chickens leading to substantial production losses, where the incidence reaches 12% in commercial chickens. It occurs primarily in heavier chickens being a modification of the breast muscle characterized by the presence of pale parallel streaks in the same orientation of the muscle fibers. Since the WS etiology remains unclear, we aimed to identify the biological and genetic mechanisms involved in its occurrence through the whole transcriptome analysis of WS in affected and unaffected chicken breast muscles. A total of 11,177 genes were expressed in the pectoralis major muscle. Out of those, 1,441 genes were differentially expressed (FDR ≤ 0.01) between the two analyzed groups, being, respectively, 772 genes upregulated and 669 downregulated in the WS affected group. A total of 36 significantly overrepresented GO terms related to WS myopathy were enriched, and the most relevant biological processes were activation of immune system, angiogenesis, hypoxia, cell death, and striated muscle contraction. The unbalance of those biological processes may trigger the occurrence of the WS phenotype in broilers. The possible lack of capillary blood supply homogeneously in the muscle triggers the hypoxia, following the activation of glycolysis, calcium signaling and apoptosis related genes facilitating the tissue damage and WS incidence.

Relationship of runs of homozygosity with adaptive and production traits in a paternal broiler line.

Genomic regions under high selective pressure present specific runs of homozygosity (ROH), which provide valuable information on the genetic mechanisms underlying the adaptation to environment imposed challenges. In broiler chickens, the adaptation to conventional production systems in tropical environments lead the animals with favorable genotypes to be naturally selected, increasing the frequency of these alleles in the next generations. In this study, ~1400 chickens from a paternal broiler line were genotyped with the 600 K Affymetrix® Axiom® high-density (HD) genotyping array for estimation of linkage disequilibrium (LD), effective population size (N e ), inbreeding and ROH. The average LD between adjacent single nucleotide polymorphisms (SNPs) in all autosomes was 0.37, and the LD decay was higher in microchromosomes followed by intermediate and macrochromosomes. The N e of the ancestral population was high and declined over time maintaining a sufficient number of animals to keep the inbreeding coefficient of this population at low levels. The ROH analysis revealed genomic regions that harbor genes associated with homeostasis maintenance and immune system mechanisms, which may have been selected in response to heat stress. Our results give a comprehensive insight into the relationship between shared ROH regions and putative regions related to survival and production traits in a paternal broiler line selected for over 20 years. These findings contribute to the understanding of the effects of environmental and artificial selection in shaping the distribution of functional variants in the chicken genome.

Association of RUNX2 and TNFSF11 genes with production traits in a paternal broiler line.

Intense selection for production traits has improved the genetic gain of important economic traits. However, selection for performance and carcass traits has led to the onset of locomotors problems and decreasing bone strength in broilers. Thus, genes associated with bone integrity traits have become candidates for genetic studies in order to reduce the impact of bone disorders in broilers. This study investigated the association of the RUNX2 and TNFSF11 genes with 79 traits related to performance, carcass composition, organs, and bone integrity in a paternal broiler line. Analyses of genetic association between single-nucleotide polymorphisms (SNPs) and traits were carried out using the maximum likelihood procedures for mixed models. Genetic associations (P < 0.05) were found between SNP g.124,883A>G in the RUNX2 gene and chilled femur weight (additive plus dominance deviation effects within sex) and with performance traits (additive within sex and additive effects). The SNP g.14,862T>C in the TNFSF11 gene presented genetic associations (P < 0.05) with additive plus dominance deviation effects within sex for performance traits. Suggestive genetic associations (P < 0.10) were found with abdominal fat and its yield. Selection based on SNPs g.14,862T>C in TNFSF11 and g.124,883A>G in RUNX2 could be used to improve performance and carcass quality traits in the population studied, although SNP g.14,862T>C was not in Hardy-Weinberg equilibrium because it was not undergoing a selection process. Furthermore, it is important to validate these markers in an unrelated population for use in the selection process.

The involvement of RUNX2 and SPARC genes in the bacterial chondronecrosis with osteomyelitis in broilers.

Economic losses due to an increase of leg disorders in broilers have become a major concern of the poultry industry. Despite the efforts to reduce skeletal abnormalities in chickens, insufficient progress has been made. Bacterial chondronecrosis with osteomyelitis (BCO) is one of the main disorders that affect bone integrity in broilers. However, the genetic pathways and genes involved in most bone problems, including BCO, remains unclear. In this study, femoral samples from male broilers with 45 days of age affected or not with BCO were used to compare the relative expression with a reverse transcription real time PCR approach of 13 candidate genes: SPP1 (osteopontin), TNFRSF11B (osteoprotegerin), SPARC (osteonectin), CALB1 (calbidin 1), CALM (Calmodulin 2), IBSP (sialoprotein), COL1A2 (collagen, type I, α 2), BMP2 (bone morphogenetic protein 2), BMP3 (bone morphogenetic protein 3), RANKL (κ-B nuclear factor ligand), SMAD1 (SMAD family member 1), LEPR (leptin receptor) and RUNX2 (related transcription factor Runt 2). Differential expression test between affected and non-affected groups was performed using the REST software. The RUNX2 and SPARC genes were downregulated (P<0.05) in the affected group, with reduced expression of fourfold when compared with the non-affected group. This result indicates that the downregulation of RUNX2 and SPARC can contribute to an increased incidence of BCO in broilers.

Genome-wide association study of periweaning failure-to-thrive syndrome (PFTS) in pigs.

Porcine periweaning-failure-to-thrive syndrome (PFTS) is a condition that affects newly weaned piglets. It is characterised by a progressive debilitation leading to death, in the absence of infectious, nutritional, management or environmental factors. In this study, we present the first report of PFTS in South America and the results of a genome-wide association study to identify the genetic markers associated with the appearance of this condition in a crossbred swine population. Four chromosomal regions were associated with PFTS predisposition, one located on SSCX, one on SSC8, and the two other regions on SSC14. Regions on SSC8 and SSC14 harbour important functional candidate genes involved in human depression and might have an important role in PFTS. Our findings contribute to the increasing knowledge about this syndrome, which has been investigated since 2007, and to the identification of the aetiology of this disease.

Comparison of iterated single-step and Bayesian regressions on genomic evaluations for age at 100 kg in swine.

The effects of modified single-step genomic best linear unbiased prediction (ssGBLUP) iterations on GEBV and SNP were investigated using 85,388 age at 100 kg phenotypes from the BRF SA breeding program Landrace pure line animals, off-tested between 2002 and 2013. Pedigree data comprised animals born between 1999 and 2013. A total of 1,068 animals were assigned to the training population, in which all of them had genotypes, original and corrected age at 100 kg phenotypes, and weighted deregressed proof records. A total of 100 genotyped animals, with high accuracy age at 100 kg estimated breeding values, were assigned to the validation population. After applying the quality control workflow, a set of 41,042 SNP was used for the analysis. Standard and modified ssGBLUP, BayesCπ, and Bayesian Lasso were compared, and their predictive abilities were accessed by approximate true and GEBV correlations. Modified ssGBLUP iteration effects on SNP estimates and GEBV were relevant, in which assigned differential weights and shrinkage caused important losses on ssGBLUP predictive ability for age at 100 kg GEBV. Even though ssGBLUP accuracy can be equal or better than the compared Bayesian methods, additional gains can be obtained by correctly identifying the number of iterations required for best ssGBLUP performance.

Association between ACTA1 candidate gene and performance, organs and carcass traits in broilers.

This study investigates the genetic association of the SNP present in the ACTA1 gene with performance traits, organs and carcass of broilers to help marker-assisted selection of a paternal broiler line (TT) from EMBRAPA Swine and Poultry, Brazil. Genetic and phenotypic data of 1,400 broilers for 68 traits related to body performance, organ weights, weight of carcass parts, and yields as a percentage of organs and carcass parts were used. The maximum likelihood method, considering 4 analytical models, was used to analyze the genetic association between the SNP and these important economic traits. The association analysis was performed using a mixed animal model including the random effect of the animal (polygenic), and the fixed effects of sex (2 levels), hatch (5 levels) and SNP (3 levels), besides the random error. The traits significantly associated (P<0.05) with the SNP were analyzed, along with body weight at 42 days of age (BW42), by the restricted maximum likelihood method using the multi-trait animal model to estimate genetic parameters. The analysis included the residual and additive genetic random effects and the sex-hatch fixed effect. The additive effects of the SNP were associated with breast meat (BMY), liver yield (LIVY), body weight at 35 days of age (BW35); drumstick skin (DSW), drumstick (DW) and breast (BW) weights. The heritability estimates for these traits, in addition to BW42, ranged from 0.24±0.06 to 0.45±0.08 for LIVY and BW35, respectively. The genetic correlation ranged from 0.02±0.18 for LIVY and BMY to 0.97±0.01 for BW35 and BW42. Based on the results of this study, it can be concluded that ACTA1 gene is associated with performance traits BW35, LIV and BMY, DW, BW and DW adjusted for body weight at 42 days of age. Therefore, the ACTA1 gene is an important molecular marker that could be used together with others already described to increase the economically important traits in broilers.

Phenotypic, genetic and environmental parameters for traits related to femur bone integrity and body weight at 42 days of age in a broiler population.

Intense selection among broilers, especially for performance and carcass traits, currently favors locomotion problems and bone resistance. Conducting studies relating to development and growth of bone tissue in broilers is necessary to minimize losses. Thus, genetic parameters were estimated for a broiler population's phenotypic traits such as BW at 42 d of age (BW42), chilled femur weight (CFW) and its yield (CFY), and femur measurements: calcium, DM, magnesium, phosphorus, and zinc content; breaking strength; rigidity; length; and thickness. Variance components were estimated through multitrait analyses using the restricted maximum likelihood method. The model included a fixed group effect (sex and hatch) and additive and residual genetic random effects. The heritability estimates we obtained ranged from 0.10 ± 0.05 to 0.50 ± 0.08 for chilled femur yield and BW42, respectively, and indicated that the traits can respond to the selection process, except for CFY, which presented low-magnitude heritability coefficients. Genetic correlation estimates between breaking strength, rigidity, and traits related to mineral content indicated that selection that aims to improve the breaking strength resistance of the femur is highly correlated with mineral content. Given the genetic correlation estimates between BW42 and minerals, it is suggested that in this population, selection for BW42 can be performed with greater intensity without affecting femoral integrity.

Identification of polymorphisms associated with production traits on chicken (Gallus gallus) chromosome 4.

Genetic selection for production traits has resulted in a rapid improvement in animal performance and development. Previous studies have mapped quantitative trait loci for body weight at 35 and 41 days, and drum and thigh yield, onto chicken chromosome 4. We investigated this region for single nucleotide polymorphisms and their associations with important economic traits. Three positional candidate genes were studied: KLF3 (Krüeppel-like factor 3), SLIT2 (Slit homolog 2), and PPARGC1A (peroxisome proliferator-activated receptor gamma, coactivator 1 alpha). Fragment sequencing of these genes was conducted in 11 F1 animals, and one polymorphism in each gene was selected and genotyped in an F2 population (N = 276) and a paternal broiler line TT (N = 840). Associations were identified with growth, carcass, and fat traits in the F2 and the paternal line (P < 0.05). Using single markers in both the F2 and the TT line, KLF3 was associated with weight gain (P < 0.05), PPPARGC1A was associated with liver and wing-parts weights and yields (P < 0.05), and SLIT2 was associated with back yield (P < 0.05) and fat traits (P < 0.05). Using multiple markers, KLF3 lost its significance in both populations, and SLIT2 was associated with feed conversion only in the TT population (P < 0.05). The QTLs mapped in the F2 population could be partly explained by PPARGC1A and SLIT2, which were associated with body weight at 35 and 41 days, respectively, and with drum and thigh yield in the same population. The results of this study indicate the importance of these genes for production traits.

Genetic parameters and mapping quantitative trait loci associated with tibia traits in broilers.

Selection among broilers for performance traits is resulting in locomotion problems and bone disorders, once skeletal structure is not strong enough to support body weight in broilers with high growth rates. In this study, genetic parameters were estimated for body weight at 42 days of age (BW42), and tibia traits (length, width, and weight) in a population of broiler chickens. Quantitative trait loci (QTL) were identified for tibia traits to expand our knowledge of the genetic architecture of the broiler population. Genetic correlations ranged from 0.56 ± 0.18 (between tibia length and BW42) to 0.89 ± 0.06 (between tibia width and weight), suggesting that these traits are either controlled by pleiotropic genes or by genes that are in linkage disequilibrium. For QTL mapping, the genome was scanned with 127 microsatellites, representing a coverage of 2630 cM. Eight QTL were mapped on Gallus gallus chromosomes (GGA): GGA1, GGA4, GGA6, GGA13, and GGA24. The QTL regions for tibia length and weight were mapped on GGA1, between LEI0079 and MCW145 markers. The gene DACH1 is located in this region; this gene acts to form the apical ectodermal ridge, responsible for limb development. Body weight at 42 days of age was included in the model as a covariate for selection effect of bone traits. Two QTL were found for tibia weight on GGA2 and GGA4, and one for tibia width on GGA3. Information originating from these QTL will assist in the search for candidate genes for these bone traits in future studies.

Genetic and phenotypic parameters of carcass and organ traits of broiler chickens.

The objective of this study was to estimate the genetic and environmental parameters for carcass, carcass part, and organ weights in a paternal strain of broiler chickens that was selected mainly for body weight at 42 days of age (BW42) to provide support for poultry genetic improvement programs. A total of 1448 chickens were used that resulted from the expansion of a pure paternal strain named TT, which was developed by Embrapa Suínos e Aves. The following weights were evaluated: BW42, chilled carcass, wing, drumstick meat, thigh meat, breast meat, breast fillet, back, liver, heart (HRT), and gizzard (GIZ). The variance component was estimated by the restricted maximum likelihood method using a multi-trait animal model. The general model included the additive genetic and residual random effects and the fixed effect of the sex-hatch group (10 levels). The heritability estimates ranged from 0.27 ± 0.06 for HRT to 0.44 ± 0.08 for GIZ. These results indicated that all the traits have enough additive genetic variability to respond to selection. The genetic correlation estimates between BW42 and the carcass and carcass part weights were high and positive. However, the genetic correlation estimates between BW42 and organ weights were low. In this population, the carcass traits might respond indirectly to selection applied to BW42. It can be concluded that selection to increase BW42 is not effective in improving broiler organ weight. Therefore, to obtain suitable genetic improvement for these traits, the selection indexes for broilers should include organ weight-based criteria.

Transcript profiling of expressed sequence tags from semimembranosus muscle of commercial and naturalized pig breeds.

In general, genetic differences across different breeds of pig lead to variation in mature body size and slaughter age. The Commercial breeds Duroc and Large White and the local Brazilian breed Piau are ostensibly distinct in terms of growth and muscularity, commercial breeds are much leaner while local breeds grow much slower and are fat type pigs. However, the genetic factors that underlie such distinctions remain unclear. We used expressed sequence tags (ESTs) to characterize and compare transcript profiles in the semimembranosus muscle of these pig breeds. Our aim was to identify differences in breed-related gene expression that might influence growth performance and meat quality. We constructed three non-normalized cDNA libraries from semimembranosus muscle, using two samples from each one, of these three breeds; 6902 high-quality ESTs were obtained. Cluster analysis was performed and these sequences were clustered into 3670 unique sequences; 24.7% of the sequences were categorized as contigs and 75.3% of the sequences were singletons. Based on homology searches against the SwissProt protein database, we were able to assign a putative protein identity to only 1050 unique sequences. Among these, 58.5% were full-length protein sequences and 17.2% were pig-specific sequences. Muscle structural and cytoskeletal proteins, such as actin, and myosin, were the most abundant transcripts (16.7%) followed by those related to mitochondrial function (12.9%), and ribosomal proteins (12.4%). Furthermore, ESTs generated in this study provide a rich source for identification of novel genes and for the comparative analysis of gene expression patterns in divergent pig breeds.

Expressed sequenced tags profiling of resistant and susceptible Gyr x Holstein cattle infested with the tick Rhipicephalus (Boophilus) microplus.

Tick resistance in cattle is mainly found in zebu (Bos indicus) animals, although it is also present in some taurine (B. taurus) breeds. In order to characterize functional genes involved in tick resistance/susceptibility in cattle, two cDNA libraries were generated using skin tissues of selected Holstein x Gyr animals. A total of 2700 high-quality reads from both resistant and susceptible cDNA were assembled into 458 sequences (contigs) and 834 singletons, with a mean size of 447.7 nucleotides. Assignment of homologous proteins by BLASTX revealed 790 (61.1%) and 300 (23.2%) hits in resistant and susceptible cDNA, respectively; 121 of these hits matched bovine proteins. A total of 502 (38.9%) unique sequences were found to have no significant homology with known sequences and were classified as novel sequences. In general, the most abundant sequences consisted of those coding for hypothetical proteins whose function had not yet been determined, in addition to ribosomal proteins, binding proteins and structural proteins, such as keratin and collagen. The most abundant protein found was collagen type III alpha, although ribosomal proteins accounted for half of the 40 most frequent hits. In addition, five matches within the top 40 best hits corresponded to immune response proteins. These sequences could be used for future studies on functional genomics of cattle tick resistance as well as for genomic sequencing projects.

Differential expression of genes in resistant versus susceptible Gyr x Holstein cattle challenged with the tick Rhipicephalus (Boophilus) microplus.

The bovine tick Rhipicephalus (Boophilus) microplus causes major losses in cattle production systems in tropical regions. Bos indicus breeds are more resistant to ticks than B. taurus breeds. Resistance genes could be an alternative to control this parasite. We examined the pattern of gene expression of three calcium-binding-protein genes: translationally controlled tumor protein 1 (TPT1), allergen Bos d3 (S100A7), calcium channel protein transient receptor potential vanilloid 6 (TRPV6), and the cysteine proteinase inhibitor gene (CST6). These genes were selected from cDNA libraries prepared from skin biopsies taken from resistant and susceptible Gyr x Holstein F2 animals. These biopsies were also used to study the expression level of these genes through real-time PCR analysis. The relative expression levels of the S100A7, TPT1, TRPV6, and CST6 genes were 2.01 ± 0.6, 1.32 ± 0.9, 1.53 ± 1.2, and 2.03 ± 0.7 times higher in the susceptible group, respectively. Skin lesion tissue from the susceptible animals showed significantly more mRNA transcripts of these genes in comparison with the resistant animals (P = 0.001). However, this hypersensitivity does not seem to protect the susceptible animals against tick infestation.

Mapeamento de locos de características quantitativas associados à composição de carcaça, no cromossomo seis de suíno / Mapping of quantitative trait loci for carcass composition on swine chromosome six

Uma população de suínos, composta de 550 animais F2, foi produzida a partir do intercruzamento da geração F1, obtida pelo cruzamento divergente de dois machos da raça nativa brasileira Piau com 18 fêmeas comerciais. O objetivo do trabalho foi mapear locos de características quantitativas (QTL) associados a cortes de carcaça. Os animais foram genotipados para 13 marcadores microssatélites, distribuídos no cromossomo 6 de suínos. As características avaliadas foram: peso total do pernil, peso do pernil sem pele e sem capa de gordura, peso total da copa, peso da copa sem pele e sem capa de gordura, peso total da paleta, peso da paleta sem pele e sem capa de gordura, peso total do carré, peso do lombo, peso total do bacon, peso das costelas, peso total da papada, peso do filezinho e peso da banha rama. Utilizou-se o método de regressão por intervalo de mapeamento por meio do programa QTL Express. Foram encontrados indicativos de QTL para peso de pernil limpo, peso de paleta, peso de lombo e peso de filezinho. A região genômica deve ser saturada com marcadores adicionais para confirmar a presença de QTL reais.

A swine population of 550 F2 animals was produced by outbred cross using two sires of the native Brazilian breed Piau and 18 commercial dams. The animals were genotyped for 13 microsatellite markers. The evaluated composition traits of carcass were: ham weight, skinless and fatless ham weight, boston shoulder weight, skinless and fatless boston shoulder weight, picnic shoulder weight, skinless and fatless picnic shoulder weight, total loin (bone-in) weight, loin weight, bacon weight, rib weight, jowl weight, sirloin weight, and belly fat weight. Data were analyzed by multiple regression interval mapping, using the QTL Express software. Suggestive QTL were found for skinless and fatless ham weight, picnic shoulder weight, loin weight, and sirloin weight. However, the genomic region should be saturated with additional markers in order to confirm the presence of real QTL.

Novos polimorfismos no gene da obesidade em raças divergentes de suínos / Polymorphisms in the leptin gene in divergent swine breeds

Investigou-se a existência de polimorfismo no gene da leptina (gene da obesidade) entre varrões da raça nativa Piau (porco tipo banha) e matrizes mestiças de raças comerciais (Landrace/Large White e Landrace/Large White com Pietrain), selecionadas para peso e precocidade. Oito pares de primers foram desenhados a partir da seqüência disponível no GenBank (U66254), usada, neste trabalho, como seqüência de referência. Amostras de DNA foram extraídas de células sangüíneas brancas utilizando-se solução de fenol:clorofórmio, após tratamento com proteinase K. Os fragmentos gerados por amplificação da reação em cadeia da polimerase foram purificados e seqüenciados em seqüenciador automático. As seqüências de nucleotídeos, obtidas a partir do DNA das raças comerciais de suíno, apresentaram maior similaridade com a seqüência de referência, e as seqüências geradas a partir do DNA dos animais nativos divergiram de ambas em algumas posições. Dos 28 polimorfismos encontrados, oito foram observados em apenas uma das três seqüências geradas a partir do DNA das raças nativas. Doze estavam presentes em duas seqüências, e os oito polimorfismos restantes foram encontrados nos três animais nativos.

Leptin gene (obese gene) polymorphism was investigated in Piau boars (a fat, native breed) and sows from commercial strains (Landrace/Large White and Landrace/Large White by Pietrain) chosen for rapid growth and early sexual maturity. Eight pairs of primers designed using the sequence available from GenBank (access n° U66254) were identified as the reference sequence in this project. DNA samples were extracted from white blood cells using phenol:chloroform solution, after treatment with proteinase K. Fragments generated by amplification of the Polymerase Chain Reaction were purified and sequenced in an automatic sequencer. Nucleotide sequences obtained from DNA of commercial swine breeds were similar to the reference sequence; whereas sequences generated from native breed DNA diverged from the reference sequence and from domestic breed DNA. Of the 28 polymorphisms found, eight were observed in only one of the three sequences generated from DNA of native breeds. Twelve polymorphisms were present in two sequences and the eight remaining polymorphisms were found in all three categories of DNA.

Avaliaçäo de frangos de corte utilizando técnicas de análise multivariada: I - Características de carcaça / Carcass traits multivariate analysis techniques in broilers evaluation

Foram utilizadas técnicas de análise multivariada para comparar dois híbridos obtidos de linhas de frango de corte desenvolvidas pela Universidade Federal de Viçosa com dois híbridos comerciais. As características avaliadas foram: peso ao abate (PAB), peso da carcaça (PC), peso do peito (PP), peso da contra-coxa (PCC), peso da coxa (PCX) e rendimento de carcaça (RCA). A análise de variância multivariada indicou diferença significativa entre os vetores de médias das características. Pelo teste de Roy, observou-se que para PC e PP os produtos comerciais foram superiores e para PCX e RCA näo houve diferenças significativas entre os produtos comercias e os cruzamentos. Quanto à funçäo discriminante linear de Fisher, observou-se superioridade significativa dos produtos comerciais pelo teste de Roy

Estudo da divergência genética entre linhagem de matrizes de frangos de corte por meio de análise multivariada / Study on genetic divergence among broilers lines by multivariate analysis

A divergência genética entre quatro linhas de matrizes de frango de corte (L1, L2, L3 e L4) foi estudada usando métodos de análise multivariada. Foram avaliadas quatro características de importância econômica, em três períodos (inicial, médio e total): idade ao primeiro ovo, número de ovos, peso médio do ovo e peso corporal da matriz. O estudo da divergência genética entre as linhas foi feito por meio da análise de variância multivariada, análise de agrupamento usando a distância generalizada de Mahalanobis e o método de otimizaçäo de Tocher, e análise por meio de variáveis canônicas. Verificaram-se diferenças significativas pelo teste do maior autovalor de Roy entre os vetores de médias das linhas avaliadas nos três períodos. A análise de agrupamento nos períodos inicial (L1 e L2; L3 e L4) e médio (L2 e L4; L1 e L3) agrupou os quatro genótipos em três grupos, No período total os genótipos foram agrupados em apenas dois grupos (L2, L3, L4 e L1). Na análise por meio de variáveis canônicas verificou-se que as duas primeiras variáveis canônicas explicaram mais de 96 por cento da variaçäo total nos três períodos, o que possibilitou representar o comportamento dos genótipos em um gráfico de dispersäo bidimensional. Observou-se que a divergência genética diminuiu quando se considerou o período total de produçäo