基于SNP芯片的猪基因组选择效果分析

为探究单步基因组最佳线性无偏预测(SSGBLUP)法应用于生猪育种的选择效果，选取杜洛克、长白、大白种猪共1 996头，利用SNP芯片获得个体基因型数据，结合表型数据和系谱数据，利用HIBLUP软件的SSGBLUP模型和基于系谱的最佳线性无偏预测(PBLUP)模型分别计算估计育种值，参考全国种猪遗传评估中心的标准计算综合选择指数，利用理论准确性和后裔测定成绩评估选择效果。结果表明：通过SSGBLUP法计算达100 kg日龄、达100 kg背膘厚及总产仔数的基因组估计育种值(GEBV)与PBLUP法计算的估计育种值(EBV)的相关系数均大于0.8;达100 kg日龄、达100 kg背膘厚及总产仔数GEBV准确性相对于EBV准确性均有所提高；SSGBLUP法与PBLUP法选留的长白和大白种猪，其后代的生长速度显著优于场内选择法选留种猪的后代。本试验中，SSGBLUP法与PBLUP法均能有效提高选留种猪后代的生长速度且二者分别计算的GEBV和EBV相关性高，但SSGBLUP法选种的准确性更高，后期可利用全基因组选择对场内现有种猪进行选种指导。     

大白母系种猪优良性状的选育进展

组建8个独立血统10头公猪、50头母猪含2个品系的大白核心选育群,通过场内种猪测定,应用GBS育种软件估计单项育种值,根据综合选择指数大小进行选留。经过4个年份的选育,年份平均校正增重公猪提高7.90%,母猪提高7.41%,均达到差异显著(P<0.05),达100 kg体重日龄缩短了1.176 d;平均校正背膘厚公猪下降了9.54%,母猪下降了7.54%,均达到差异显著(P<0.05%),达100 kg体重背膘厚育种值减少0.159 mm,母猪总产仔数逐年提高,产仔数育种值增加0.029头;综合选择指数提高3.84%。选育结果表明,组建优秀的核心选育群,通过不同品系间的杂交,能够培育出高性能大白专门化母系种猪。     

我国荷斯坦青年公牛基因组选择效果分析

本研究基于我国荷斯坦奶牛基因组遗传评估和生产性能测定（DHI）结果,旨在分析我国荷斯坦公牛基因组选择的效果。选择1 686头既有基因组遗传评估成绩又有后裔测定成绩的荷斯坦公牛,利用2019年12月基因组遗传评估结果及其女儿的产奶和体型性状数据,通过R软件与Excel计算公牛基因组评估结果与公牛女儿表型数据间的相关性,对我国荷斯坦青年公牛基因组选择效果进行分析。相关性分析结果表明,荷斯坦公牛的基因组性能指数（GCPI）与后裔测定性能指数（CPI）呈正相关（r_s>0.3）,其中产奶量和体细胞评分的基因组育种值（GEBV）与估计育种值（EBV）呈较强的正相关（0.4 < r_s < 0.8）。对公牛女儿表型数据分析结果表明,女儿产奶量、乳蛋白率、乳脂率与肢蹄评分的表型值与公牛GEBV分组趋势一致,且公牛不同产奶性状GEBV组间的女儿性状表型值大部分达到极显著差异（P<0.01）;北京及上海地区公牛产奶性状、体细胞评分和肢蹄评分的GEBV分组与女儿表型值趋势较其他省市（地区）更一致,且GEBV高组与低组之间差值均高于其他省市（地区）。基于1 686头荷斯坦公牛基因组选择及其女儿表型数据的分析结果表明,我国荷斯坦公牛的基因组遗传评估准确性较好,其中产奶量、乳蛋白率、体细胞评分和肢蹄评分的表型数据更好地反映了基因组选择的效果;北京及上海地区较其他省市（地区）更能反映我国荷斯坦公牛基因组选择的效果。     

北京市种猪基因组选择育种平台的构建与应用

2006年,北京市建立了种猪遗传评估数据平台。截止到2021年底,平台共收录371.46万头种猪系谱信息,51.23万条达100 kg体重日龄记录,50.90万条达100 kg活体背膘厚记录和45.15万条窝产仔记录。2015年,北京市将常规育种与基因组育种相结合,启动了种猪基因组选择育种平台构建工作,北京市种猪遗传评估进入基因组选择时代。截止到2021年底,平台建立了大白猪基因组选择参考群体,规模达到5 335头;开发基因组遗传评估系统,实现了将复杂的基因组选择计算过程转化为“一键式”操作;利用一步法（ssGBLUP）对9个种猪场的4 631头大白猪进行基因组遗传评估。基因组选择实施后,选择准确性大幅提高。早期选择时,达100 kg体重日龄的育种值准确性由0.56提高至0.66,达100 kg活体背膘厚的基因组育种值准确性由0.56提高至0.70,总产仔数育种值准确性由0.41提高至0.60。终选时,达100 kg体重日龄的育种值准确性由0.69提高至0.79,达100 kg活体背膘厚的基因组育种值准确性由0.72提高至0.80,总产仔数性状育种值准确性由0.41提高至0.61。基因...     

湘沙猪配套系选育及展望

湘沙猪配套系以地方良种沙子岭猪和引进品种巴克夏、大白猪为育种素材,经过5个世代的持续选育,分别育成XS1系(终端父本)、XS2系(母系父本)、XS3系(母系母本)3个专门化品系。各专门化品系主要经济技术指标达到或超过育种目标,遗传稳定。XS1系初产总仔数10.7头,产活仔数10.3头;经产总产仔数11.8头,产活仔数11.4头;公猪达100 kg日龄165.4 d,背膘厚9.9 mm;母猪达100 kg日龄166.5 d,背膘厚10.8 mm。XS2系初产总仔数10.0头,产活仔数9.5头;经产总产仔数10.2头,产活仔数9.8头;公猪达100 kg日龄168.8 d,背膘厚12.5 mm;母猪达100 kg日龄169.7 d,背膘厚13.5 mm。XS3系初产总仔数10.2头,产活仔数9.4头;经产总产仔数11.1头,产活仔数10.3头;公猪达50 kg日龄193.9 d,背膘厚15.6 mm;母猪达50 kg日龄185.7 d,背膘厚17.9 mm。配套系父母代总产仔数12.4头,产活仔数11.9头。配套系商品猪30~100 kg期间平均日增重832.44 g,料重比3.16;100.6 kg屠宰,屠宰率73.0%,胴体瘦肉率58.2%,系水力93.26%,肌内脂肪含量2.9%,肉质优良,市场前景广阔。     

克隆杜洛克种公猪及其扩繁后代的生产性能评价

为评估克隆种公猪在生猪产业上的应用价值,进行杜洛克种公猪的克隆及扩繁试验研究。采集2头常规选育的杜洛克种公猪耳组织,建立细胞系、生产重构胚,移植到受体母猪,分娩克隆公猪并调教采精,采集精液并参配纯种杜洛克母猪,统计其生产性能,以同期普通杜洛克公猪及其后代为对照。结果显示：（1）建立了种猪耳组织细胞建系、重构胚生产、胚胎移植等技术体系;（2）移植7头代孕母猪,共产仔39头克隆杜洛克公猪,断奶存活21头,60日龄存活17头,达100 kg体重日龄测定15头,调教采精成功10头,种猪利用率为25.6%（10/39）;（3）获得的克隆公猪与同期杜洛克公猪在初生重、断奶重、校正100 kg体重日龄、校正背膘厚等指标上无显著性差异;（4）获得的59头杜洛克克隆公猪后代与同期普通杜洛克公猪后代在初生重、断奶重、校正100 kg体重日龄、校正背膘等指标上无显著性差异,仅在右乳头数指标上,克隆公猪后代（6.34±0.48）显著低于普通公猪后代（6.60±0.66）。结论：建立了种猪克隆快速扩繁技术平台,但常规选育的克隆公猪及其后代生产性能与普通公猪及其后代无显著差异,需要进一步提高选择准确性,同时提高克隆效率及其存活率,才能满足克隆技术应用于终端公猪生产的产业要求。     

杜洛克猪生长性状全基因组关联分析及候选基因鉴定

旨在对杜洛克猪生长性状进行全基因组关联分析及候选基因鉴定。本研究选用361头杜洛克种公猪作为试验群体，对达100 kg体重日龄、达100 kg平均日增重、达100 kg活体背膘厚和达100 kg眼肌面积性状进行测定，基因型信息使用50K单核苷酸多态性阵列进行分型，质控后得到31 618个SNPs。使用GCTA软件利用基因组信息对各生长性状进行遗传参数估计，使用R软件rMVP包FarmCPU模型进行全基因组关联分析，鉴定与生长性状相关的基因组区域和候选基因。结果表明，达100 kg体重日龄、达100 kg平均日增重、达100 kg活体背膘厚和达100 kg眼肌面积性状的遗传力分别为0.27、0.29、0.16和0.11,属于中等遗传力性状，达100 kg体重日龄和达100 kg平均日增重的遗传相关和表型相关值均为-0.99,为强负相关关系。全基因关联分析结果表明，在达100 kg体重日龄和达100 kg平均日增重性状上共检测到3个显著SNPs,均位于10号染色体上。使用最小显著差数检验法对显著SNPs的等位基因型进行多重比较，显著SNPs rs81237156、rs81424502和rs...     

不同近交程度对大白猪生长性能的影响

遗传学上将有亲缘关系的个体间的交配称为近交,但在育种学上认为,如果双方间存在亲缘关系就叫近亲交配,简称近交。本文使用stata软件通过单因素方差分析研究了不同近交程度对大白猪背膘厚的表型值、EBV值和大白猪达100 kg体重的日龄的表型值、EBV值的影响。试验结果表明:近交系数低于12.5%时对背膘厚度和达100 kg日龄的表型值、EBV值的影响均不显著(P》0.05);近交系数高于1 2.5鬈时仅对达100 kg日龄EBV值的影响有极显著差异(P≤0.01),对其他三个性状影响均不显著(P>0.05)。可见,温和近交(近交系数<12.5%)对大白品种的猪背膘厚和达100 kg体重日龄的影响不显著。     

猪体型评分性状的相关性分析

为了分析环山集团近2年来某核心育种场猪体型评分性状与其他生产性状之间的相关性,笔者利用R软件评估了总体评分与各部位评分以及与繁殖性状的表型相关,又利用DMU软件和动物模型估计了体型评分性状与达100 kg体重日龄和背膘厚之间的表型与遗传相关。结果表明:总体评分与四肢、生殖器官和中躯表型相关较大,分别达到0.597,0.481和0.432;与总产仔数和产活仔数表型相关接近于0,分别达到0.041和0.034;此外,体型评分与达100 kg体重日龄和背膘厚表型相关也接近于0,而遗传相关分别达到-0.16和0.05。说明体型评分还需要进一步优化,在父系和母系中应选择性应用。     

不同性别商品猪体尺性状、胴体性能及肉质性状比较

选择体重100 kg左右的商品猪100头,阉公猪去势,母猪不去势,研究宰前体尺性状、活膘及宰后胴体性能和肉品质。结果表明:倒数3～4腰椎膘厚和倒数3～4胸椎膘厚,母猪组分别比阉公猪组降低5.97%(P>0.05)和7.32%(P>0.05);倒数3～4胸椎肌肉厚度,母猪组比阉公猪组提高7.09%(P>0.05);活体瘦肉率,母猪组比阉公猪组提高2.67%(P>0.05);皮脂率,母猪组比阉公猪组降低3.39%(P>0.05);瘦肉率,母猪组比阉公猪组提高1.36%(P>0.05);肉质性状,母猪组与阉公猪组差异不显著。     

Using Progeny Testing to Evaluate the Efficiency of Genomic Selection in Large White Pigs

This study aimed to evaluate the actual genetic improvement effect of genomic selection in Large White boars through progeny testing in production performance. Nine hundred and thirteen Large White pigs were used to construct a reference group, and 823 new-born Large White boars were used to implement the first genomic selection through ssGBLUP before castration. The second genomic selection were carried out after performance testing, then 10 boars with significant difference in production performance were selected and their offsprings were compared in phenotypic values, estimated breeding values of growth traits and selection index. The results showed that the accuracies of genomic prediction on age at 100 kg body weight, 100 kg backfat thickness and total number born increased from 0.56, 0.67 and 0.64 in the first genomic selection to 0.73, 0.73 and 0.67 in the second genomic selection, respectively. The correlation coefficient of maternal selection index between the two genomic selection before castration and after performance testing was 0.82, which indicated that the first genomic selection before castration was accurate enough to make early selection on boars. According to the genomic breeding values and maternal selection index of 10 selected boars, two groups with high and low production performance were set up. The progeny testing showed that the difference of average phenotypic value between groups was 2.58 days, and the difference of average evaluated breeding value(EBV) between groups was 3.08 days in age at 100 kg body weight, those were 1.15 mm and 1.03 mm in 100 kg backfat thickness, respectively, and the difference in the mean of the comprehensive maternal index was 9.3, all the differences(except age at 100 kg body weight) were extremely significant. This study prove that the offspring of boars with significant differences in genomic evaluation have significant differences in phenotypic values and breeding values, which indicate that, through genomic selection, excellent breeding boars can be selected and their genetic superiority can be passed to their offsprings.     

大白猪基因组选择在不同应用策略下的比较研究

目前,基因组选择（genomic selection,GS）技术已经在种猪育种中开展,但为获得较高的收益,还需研究一些应用策略,如确定仔猪基因分型个体比例和早期仔猪留种比例。本试验选择温氏集团出生于2011—2016年的大白种猪作为研究对象,共有超过4.5万条的生长测定记录,超过7万条繁殖记录,和2 090个个体的简化基因组测序（GBS）数据,其中,出生于2016年7~12月的440个体作为候选群体。研究性状包括两个生长性状（校正100 kg日龄和校正100 kg背膘厚）和一个繁殖性状（总产仔数）。为对比预测效果,在候选群体进行育种值预测时,按照是否利用其基因型或表型信息分为4种预测方案,比较不同方案的预测可靠性和个体选择指数的排名情况。结果显示,在预测候选群育种值时,利用其表型或基因型信息均比不利用时的预测结果更加可靠。对生长性状终测前、后进行基因组选择指数计算,发现,终测后指数排名前30%的个体都位于终测前指数排名前60%内。若仔猪出生后仅选择常规BLUP预测指数排名前60%的个体,会导致有接近15%的具有优秀潜力的个体被遗漏。本研究建议,对所有新生健康仔猪都进行基因分型并计算基因组选择指数,然后对指数排名靠前60%的个体进行性能测定。     

Comparative Analysis of Different Implementation Strategies on Genomic Selection in Large White Pigs

At present, genomic selection (GS) has been applied in pigs breeding, but some implementation strategies, such as the determination of genotyping ratios or early selection rates for piglets, are required to obtain a higher benefit using this technology. The Large White pigs born from 2011 to 2016 at WENS Foodstuff Group Co.,Ltd were chose as the research objects, including more than 45 000 growth measurement records, more than 70 000 reproduction records and 2 090 individuals with genotyping-by-sequencing (GBS) data. The 440 individuals born from July to December in 2016 were used as the candidate individuals. The traits included two growth traits, age at 100 kg and backfat thickness at 100 kg, and one reproduction trait, number of total born. To compare the prediction effects, four prediction scenarios were designed according to including or ignoring the phenotypic or genotypic information of candidate individuals when predicting their breeding values. The predictive reliability of different scenarios and rankings of selection indices of individuals would be compared. The results showed that the results using the phenotypic and genotypic information was more reliable than ignoring them to predict the breeding values of candidate individuals. When genomic selection indices were calculated before and after performances testing for the growth traits, the individuals ranking in the top 30% of indices after testing were all found in the individuals ranking in the top 60% of indices before testing. If the piglets with the top 60% of traditional BLUP indices were only selected, around 15% of individuals with good genetic potentials would be omitted. This study suggests that all healthy piglets after birth are genotyped and their genomic selection indices are calculated, and then the individuals ranking in the top 60% of indices are chose to perform growth measurement.     

Genetic parameters for production traits and measures of residual feed intake in large white swine.

The purpose of this study was to estimate genetic parameters for ADG, backfat thickness and loin eye area (LEA), and measures of feed intake and efficiency for purebred Large White boars born from 1990 to 1997. Boars from 60% of the litters were culled at weaning based on a maternal breeding value (index) of the dam, and remaining boars (n = 26,706) were grown to 100 d of age. Selection of boars for individual pen testing was based on a combination of growth and maternal indices. Boars were fed a corn-soybean meal diet that was 1.14% lysine, 19% protein, and 3,344 kcal/kg ME for approximately 77 d. Boars were weighed at the beginning and end of the test, and feed intake was recorded. Daily feed intake (DFI), ADG, and feed:gain ratio (FG) were computed. Four measures of residual feed intake (RFI) were estimated as the difference between actual feed intake and that predicted from models that included 1) initial test age and weight and test ADG (RFI1); 2) initial test age and weight, test ADG, and backfat (RFI2); 3) initial test age and weight, test ADG, and LEA (RFI3); and 4) initial test age and weight, test ADG, backfat, and LEA (RFI4). Genetic parameters were estimated using an animal model and single- or multiple-trait DFREML procedures. Models included fixed effects of contemporary groups and initial test age as a covariate and random animal and litter effects. Heritability estimates for test ADG, DFI, FG, backfat, LEA, RFI1, RFI2, RFI3, and RFI4 were .24, .23, .16, .36, .24, .17, .11, .15, and .10, respectively. Genetic correlations between ADG and backfat, ADG and LEA, ADG and DFI, and ADG and FG were .37, .36, .82, and -.32, respectively. Genetic correlations between ADG and measures of residual feed intake ranged from .11 to .18. Genetic correlations of backfat with LEA, DFI, and FG were -.27, .64, and .40, respectively. Genetic correlations of backfat with RFI measures were higher when backfat was not included in the estimation of RFI. Genetic correlations for LEA with DFI and FG were 0 and -.52, respectively. Genetic correlations for LEA with RFI measures were all negative and ranged from -.31 to -.51. Genetic correlations indicate that selection for reduced RFI could be made without adversely affecting ADG. Backfat should also decrease, and LEA should increase. The amount of change in backfat or LEA would depend on the measure of RFI used.     

The heritability of pH1 in longissimus dorsi muscle in landrace and large white pigs

The susceptibility of a carcass to PSE (pale, soft, exudative muscle) can be assessed by measuring the pH value in the lumbar region of the longissimus dorsi muscle at 45 min post-mortem (pH₁). The effect of breed, station, sex and season on pH₁ was investigated, on a total of 2 775 pig records, including the progeny of 129 Irish Landrace and 126 Large White boars, from two test stations. The heritability of pH₁ and its genetic correlation with other performance characteristics were determined within each breed. Landrace pigs had significantly lower values than Large White pigs, indicating a greater susceptibility to PSE in the former. There was no significant difference between the values for boars, castrates and females. There were marked differences from month to month, but no definite seasonal pattern was present nor was there any apparent long-term trend. The heritability for the Landrace breed was higher than that for the Large White breed, and both values indicate that pH₁ would respond to selection. The genetic correlations between pH₁ and daily gain, food conversion efficiency and backfat (four measurements) were for the most part low and the standard errors relatively high, indicating that there was no strong relationships between pH₁ and these performance characteristics.     

Genetic correlations between test station and on-farm performance for growth rate and backfat in pigs

Record of performance data taken on Yorkshire pigs on-farm in 123 breeder herds and at a central test station were used to estimate genetic correlations between measures of backfat depth and days to 90 kg on boars at the test station and boars and gilts on-farm. The data involved records on 3,513 station-tested boars, 13,760 farm-tested boars and 28,203 farm-tested gilts from 838, 2,098 and 2,339 sires, respectively. For backfat depth, estimates of genetic correlations were .85 for test station and on-farm boars and 1.04 for test station boars and on-farm gilts. Estimates of genetic correlations between test station and on-farm measures of days to 90 kg were .80 for boars and .74 for boars and gilts. Based on these results, selection of boars on the basis of test station performance for backfat and growth rate would be expected to result in genetic improvement on-farm in both sexes under North American testing and management conditions.     

A bioeconomic simulation model for a hierarchical swine breeding structure.

A stochastic computer model was developed to simulate individual pigs in a hierarchical breeding system. The bioeconomic model was designed as a tool to facilitate the evaluation of selection, culling, and management strategies for a three-tiered breeding structure. Events such as mating, farrowing, and selection occurred weekly. Variables included number of pigs born alive, survival rate from birth to weaning, average daily gain and backfat at 110 kg, number of pigs weaned, feed per gain, days from weaning to 110 kg, age at puberty, and growth rate and weight of sows and service boars. Also included were probabilities of conception, return to estrus by week, survival, involuntary culling, male infertility, and unacceptable conformation. Variables important for selection were determined by breeding value, individual and maternal heterosis, parity, size of birth litter, sex, age of dam, genetic and environmental relationships between variables, and common litter, permanent, and random environmental effects. Variables derived from selection variables were computed by regression using phenotypic relationships between all variables. Also, a random environmental effect was added to predicted performance. Means and variances of variables differed between genetic lines. Production costs included feed, non-feed operating, fixed, and replacement stock costs. Income included market animals, culls, and replacements sold to lower tiers. Effects of changes in backfat on market value and sow maintenance feed costs were not modeled. An example is given to illustrate model output.     

Genetic and phenotypic parameters for pig growth and body composition estimated by intraclass correlation and parent-offspring regression

Purebred Duroc and Yorkshire boars and gilts, farrowed in spring litters from 1974 through 1982 and in fall litters from 1974 through 1978, were maintained as closed select and control lines descended from the same base population. Spring-farrowed pigs were selected mainly on an index of sow productivity traits, whereas selection among fall-farrowed pigs was mainly on an index of pig performance traits. Basic traits analyzed were age of pig at 91 kg, postweaning average daily gain in weight, average backfat thickness (ABF) and longissimus muscle area (LMA), with ABF and LMA measured from ultrasonic scans at 91 kg. Also analyzed were estimated weight of trimmed wholesale lean cuts at 91 kg live weight and lean cuts growth rate from birth to 91 kg. Standardized selection differentials indicated that no significant selection pressure was applied to the four basic traits in the population. A nested analysis of variance of intraclass correlations among paternal half-sib families was computed with 1,930 gilt records, providing estimates of heritabilities and genetic and phenotypic correlations among the six traits. Also, estimates were computed for the portion of total phenotypic variance due to maternal-related covariances among littermates and the portion due to random environmental variances among individuals. In addition, estimates of the population parameters were computed from regressions of boars and gilts on sires, dams and mid-parental values with 974 boar and 1,686 gilt deviation records. Composite parameter estimates were then computed from the separate values weighted by the inverse of their standard errors.     

Genetic and phenotypic trends in Polish large white nucleus swine herds

Performance test records collected from 1978 to 1987 from on-farm tests of young Polish Large White boars from 94 herds and reproductive records of Polish Large White sows from 81 nucleus farms were used to estimate the phenotypic, environmental, and genetic trends. There were, after editing, 114,347 boar performance records and 41,080 litter records on sows. Both data sets were analyzed by use of an animal model. Estimated annual phenotypic and environmental trends were relatively large and desirable and were, respectively, .17 +/- .05, .11 +/- .05 (number born alive); .16 +/- .04, .10 +/- .04 (21-d litter size); 1.86 +/- .63, 1.43 +/- .62 (21-d litter weight, kg); 6.80 +/- .60, 6.76 +/- .72 (average daily gain, g/d); -.065 +/- .007, -.058 +/- .023 (backfat thickness, mm); -2.76 +/- .28, -2.75 +/- .29 (days to 110 kg). In contrast, all estimated genetic trends were relatively small and not always favorable. The genetic trends estimated from animal, sire and dam genetic values were, respectively, .01 +/- .01, .02 +/- .01, .01 +/- .01 (number born alive and 21-d litter size); .04 +/- .06, .10 +/- .05, .05 +/- .04 (21-d litter weight, kg); .04 +/- .04, .50 +/- .10, -.43 +/- .05 (average daily gain, g/d), -.009 +/- .001, -.015 +/- .002, -.004 +/- .0004 (backfat thickness, mm); and -.01 +/- .01, -.17 +/- .04, .19 +/- .02 (days to 110 kg). Neither examination of selection practices nor boar utilization provided an explanation for the lack of genetic progress.     

Relationship between Testosterone Responses to Administration of GnRH and Breeding Performance in Boars

The aim of the present study was to determine the relationship between peripheral testosterone responses to the administration of GnRH and breeding performance in boars. Twelve, sexually mature Large White boars were used. The first blood samples were collected at 9.00. At that time all boars received 100 μg GnRH (Ovurelin inj. A.U.V. Reanal). The second blood samples were taken 2.5 hours after the injection of GnRH. Serum was analyzed for testosterone by radioimmunoassay. Breeding performance was evaluated over a 150-day breeding season. The basal and the stimulated testosterone levels were related to the breeding performance. The correlation coefficient between breeding performance and post-GnRH peripheral testosterone concentration was significant (r = 0.61; P ≤ 0.05). This data suggests that the increase of the peripheral testosterone concentration may be one of the indicators for breeding performance in boars.