Meta-analysis and cross-validation studies of QTLs associated with reproductive traits in tropical beef cattle
Carregando...
Arquivos
Data
2018-12-07
Autores
Melo, Thaise Pinto de [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
O poder estatístico de um único estudo de associação genômica ampla (GWAS) para detectar loci de características quantitativas (QTL) é esperado ser menor do que combinando resultados de GWAS independentes de múltiplas raças, isto é, de diferentes raças, especialmente para características complexas, como as de puberdade. Utilizar populações independentes e métodos para detectar novas regiões genômicas e validar regiões previamente conhecidas como associadas com a característica de interesse em populações independentes é uma estratégia viável para identificar QTLs mais acuradamente. O objetivo deste estudo foi detectar regiões genômicas controlando características de precocidade sexual em três raças de gado de corte tropical (Nelore, Brahman e Composto Tropical - CT), utilizando duas estratégias, meta-analise utilizando diferentes raças e validação cruzada de QTL entre raças. No estudo de meta-análise as características incluídas foram idade ao primeiro parto (IPP), prenhez precoce (PP), e circunferência escrotal (CEN) medida aos 18 meses de idade no Nelore, e a idade ao surgimento do primeiro corpo lúteo (IPCL), intervalo do primeiro anestro pós-parto (IPAPP) e circunferência escrotal medida aos 18 meses de idade (CE18B) para o Brahman. A meta-análise foi realizada utilizando um método multi-característica. Um total de 108 polimorfismos de nucleotídeo único (SNP) significativos a um P-valor empírico de 1.39 × 10−5 (FDR < 0.05), foram encontrados neste estudo. Tais SNP significativos foram distribuídos em 19 entre 29 autossomos, e a maioria deles estava localizada no BTA14. No estudo de meta-analise foram identificadas cinco regiões abrigando a maioria dos SNP significativos (76%): BTA2 (5.55%) de 95 a 96 Mb, BTA4 (5.55%) de 94.1 a 94.8 Mb, BTA14 (59.26%) de 24 a 25 Mb e de 29 a 30 Mb, e BTA21 (5.55%) de 6.7 Mb a 11.4 Mb. No estudo de validação cruzada entre raças foram utilizadas as mesmas características do estudo de meta-analise, exceto a IPP, além das características habilidade de ovular antes do desmame do bezerro (AD) e CE medida aos 12 e 24 meses de idade (CE12, CE24), avaliadas no Brahman. O Nelore foi considerado a população de validação e o Brahman, a população de “Discovery”, no qual foram utilizados dois conjuntos de SNP: 1) SNP em regiões gênicas previamente associadas com a precocidade sexual no Brahman, e SNP em regiões vizinhas a estes genes a ± 250 Kb de distância (para mais ou para menos) (G1); e 2) SNP significativos de estudos de meta-análise para características de precocidade sexual em machos e fêmeas Brahman e suas regiões vizinhas a ± 250 Kb de distância (para mais ou para menos) (G2). Este procedimento foi utilizado para validar no Nelore algumas regiões genômicas que são conhecidas por afetar características de precocidade sexual em Brahman, e para encontrar novos QTL afetando estas características em ambas as raças. Para PP 144 SNP foram validados em G1 e 41 SNP foram validados em G2. Para CEN 14 SNP foram validados em G1 e 2 em G2.. Um total de 21 SNP candidatos foi detectado para EP, isto é, eles estiveram em ambos os conjuntos de SNP, G1 e G2. Estes SNP e suas regiões vizinhas validados em machos e fêmeas estiveram próximos a QTL ou genes associados com eventos reprodutivos em bovinos e outras espécies, e são, portanto regiões candidatas afetando características de precocidade sexual em Nelore e Brahman. Além disso, os resultados do estudo de meta-análise foram validados em uma terceira raça, o CT. Aqueles SNP que foram significativos na meta-analise sob um P-value empírico de 1 × 10-4 e também foram significativos (P-valor ≤ 1 × 10-3) para cada GWAS individual das características do CT, IPCL, IPAPP e CECT, ou estiveram localizadas a ± 250 Kb (para mais ou para menos) dos SNP significativos de cada GWAS foram considerados validados. Em resumo, um total de 49, 4 e 14 SNP foram validados para IPCL, IPAPP e CECT, respectivamente. De modo geral, os resultados encontrados indicam regiões genômicas candidatas controlando a precocidade sexual em bovinos Nelore, Brahman e CT, e estas regiões parecem abrigar genes / QTLs que estão controlando a puberdade entre as raças estudadas aqui, apesar das diferenças na origem destas raças e nas condições ambientais e de manejo aos quais os animais estão expostos. Em resumo nós pudemos validar e encontrar novas regiões candidatas que estão provavelmente controlando a precocidade sexual em três populações de bovinos de corte tropical. Esse resultado nos permite concluir que a estratégia de utilizar meta-analise com múltiplas raças e validação de QTL entre raças é viável para detectar novos QTL e validar regiões genômicas previamente associadas com as características de interesse.
The statistical power of an individual genome-wide association study (GWAS) to detect quantitative trait loci (QTL) is expected to be lower than combining independent GWAS results from multi-breed, i.e., from different breeds, especially for complex traits, as puberty traits. Using independent populations and methods to detect new genomic regions and validate in independent populations known regions associated with the trait of interest is a feasible strategy to identify QTLs more accurately. The aim with this study was to detect genomic regions controlling sexual precocity traits across three tropical beef cattle breeds (Nellore, Brahman and Tropical Composite - TC) by using two approaches, meta-analysis using different breeds and QTL validation across different breeds. In the meta-analysis study the traits included were age at first calving (AFC), early pregnancy (EP), and scrotal circumference (SCN) measured at 18 months of age for Nellore, and age at first corpus luteum (AGECL), first postpartum anoestrus interval (PPAI), and scrotal circumference measured at 18 months of age (SC18B) for Brahman cattle. The meta-analysis was performed using a multi-trait method. A total of 108 significant single-nucleotide polymorphisms (SNPs), at an empirical threshold P-value of 1.39 × 10−5 (FDR < 0.05), were found in this study. Those significant SNP were distributed over 19 out of 29 autosomes, and the major of them were located on BTA14. In the meta-analysis study we identified five association regions harbouring the majority of the significant SNP (76%), namely: BTA2 (5.55%) from 95 to 96 Mb, BTA4 (5.55%) from 94.1 to 94.8 Mb, BTA14 (59.26%) from 24 to 25 Mb and 29 to 30 Mb, and BTA21 (5.55%) from 6.7 Mb to 11.4 Mb. In the across-breed validation study we used the same traits of the meta-analysis study, except AFC, and we also used the traits ability to ovulate prior to weaning the calf (PW) and SC measured at 12 and 24 months of age (SC12, SC24), evaluated in Brahman. We considered Nellore as validation and Brahman as discovery population and used two SNP sets: 1) SNPs inside genes previously associated with sexual precocity in Brahman cattle, and SNPs within ± 250 Kb (upstream and downstream) neighbouring regions of those genes (G1); and 2) significant SNPs from meta-analyses studies for female and male sexual precocity traits in Brahman and their ± 250 Kb (upstream and downstream) neighbouring regions (G2). This procedure was performed in order to validate in Nellore some genomic regions that are known as affecting sexual precocity traits in Brahman, and to find new QTL affecting those traits in both breeds. For EP 144 SNP were validated in G1 and 41 SNP were validated in G2. For SCN 14 SNP were validated in G1 and 2 SNP were validated in G2. A total of 21 candidate SNP were detected for EP, i.e., they were in common between G1 and G2 SNP sets. These SNP and their neighbour regions validated in males and females were close to QTL and genes associated with reproductive events in bovine or other species. Therefore, they are candidate regions affecting sexual precocity traits in Nellore and Brahman. Also, the results of the meta-analysis study were validated in a third breed, the TC. Those SNP that were significant in the meta-analysis under an empirical P-value of 1 × 10-4 and were also significant (P-value ≤ 1 × 10-3) for each individual GWAS of TC traits, AGECL, PPAI and SCTC, or were located ± 250 Kb (up-stream or down-stream) from those significant SNP of each GWAS were considered validated. In summary, a total of 49, 4 and 14 SNP were validated for AGECL, PPAI and SC, respectively. In general, our results indicated candidate genomic regions controlling sexual precocity in Nellore, Brahman and TC cattle, and these regions seem to harbour the same genes / QTLs controlling puberty across breeds, despite differences in the breeds’ origin and in the environment and management conditions that those breeds are exposed. In summary we were able to validate and find new candidate regions that are likely controlling sexual precocity in three independent tropical beef cattle populations. This result allows us to conclude that the strategy of using multi-breed meta-analysis and across breed QTL validation is feasible to detect new QTL and to validate known genomic regions affecting the traits of interest.
The statistical power of an individual genome-wide association study (GWAS) to detect quantitative trait loci (QTL) is expected to be lower than combining independent GWAS results from multi-breed, i.e., from different breeds, especially for complex traits, as puberty traits. Using independent populations and methods to detect new genomic regions and validate in independent populations known regions associated with the trait of interest is a feasible strategy to identify QTLs more accurately. The aim with this study was to detect genomic regions controlling sexual precocity traits across three tropical beef cattle breeds (Nellore, Brahman and Tropical Composite - TC) by using two approaches, meta-analysis using different breeds and QTL validation across different breeds. In the meta-analysis study the traits included were age at first calving (AFC), early pregnancy (EP), and scrotal circumference (SCN) measured at 18 months of age for Nellore, and age at first corpus luteum (AGECL), first postpartum anoestrus interval (PPAI), and scrotal circumference measured at 18 months of age (SC18B) for Brahman cattle. The meta-analysis was performed using a multi-trait method. A total of 108 significant single-nucleotide polymorphisms (SNPs), at an empirical threshold P-value of 1.39 × 10−5 (FDR < 0.05), were found in this study. Those significant SNP were distributed over 19 out of 29 autosomes, and the major of them were located on BTA14. In the meta-analysis study we identified five association regions harbouring the majority of the significant SNP (76%), namely: BTA2 (5.55%) from 95 to 96 Mb, BTA4 (5.55%) from 94.1 to 94.8 Mb, BTA14 (59.26%) from 24 to 25 Mb and 29 to 30 Mb, and BTA21 (5.55%) from 6.7 Mb to 11.4 Mb. In the across-breed validation study we used the same traits of the meta-analysis study, except AFC, and we also used the traits ability to ovulate prior to weaning the calf (PW) and SC measured at 12 and 24 months of age (SC12, SC24), evaluated in Brahman. We considered Nellore as validation and Brahman as discovery population and used two SNP sets: 1) SNPs inside genes previously associated with sexual precocity in Brahman cattle, and SNPs within ± 250 Kb (upstream and downstream) neighbouring regions of those genes (G1); and 2) significant SNPs from meta-analyses studies for female and male sexual precocity traits in Brahman and their ± 250 Kb (upstream and downstream) neighbouring regions (G2). This procedure was performed in order to validate in Nellore some genomic regions that are known as affecting sexual precocity traits in Brahman, and to find new QTL affecting those traits in both breeds. For EP 144 SNP were validated in G1 and 41 SNP were validated in G2. For SCN 14 SNP were validated in G1 and 2 SNP were validated in G2. A total of 21 candidate SNP were detected for EP, i.e., they were in common between G1 and G2 SNP sets. These SNP and their neighbour regions validated in males and females were close to QTL and genes associated with reproductive events in bovine or other species. Therefore, they are candidate regions affecting sexual precocity traits in Nellore and Brahman. Also, the results of the meta-analysis study were validated in a third breed, the TC. Those SNP that were significant in the meta-analysis under an empirical P-value of 1 × 10-4 and were also significant (P-value ≤ 1 × 10-3) for each individual GWAS of TC traits, AGECL, PPAI and SCTC, or were located ± 250 Kb (up-stream or down-stream) from those significant SNP of each GWAS were considered validated. In summary, a total of 49, 4 and 14 SNP were validated for AGECL, PPAI and SC, respectively. In general, our results indicated candidate genomic regions controlling sexual precocity in Nellore, Brahman and TC cattle, and these regions seem to harbour the same genes / QTLs controlling puberty across breeds, despite differences in the breeds’ origin and in the environment and management conditions that those breeds are exposed. In summary we were able to validate and find new candidate regions that are likely controlling sexual precocity in three independent tropical beef cattle populations. This result allows us to conclude that the strategy of using multi-breed meta-analysis and across breed QTL validation is feasible to detect new QTL and to validate known genomic regions affecting the traits of interest.
Descrição
Palavras-chave
Brahman, GWAS, Nelore, Puberdade, QTL, SNP, Puberty, Validation