칼럼 - 미래기술축산 위한 전문 컨설팅업체 육성 필요

이득환,Lee, Deuk-Hwan 한국오리협회 2015 오리마을 Vol.148 No.-

한우의 주요 경제형질에 대한 유전모수 추정 1 . 체중 및 흉위에 대한 일반능력 및 자료의 분포 특성

이득환,오봉국,박영일,신언익,양영훈 ( D . H . Lee,B . K . Ohh,Y . I . Park,O . Y . Shin,Y . H . Yang ) 한국축산학회 1991 한국축산학회지 Vol.33 No.7

This study was carried out to seek the effects of environmental factors and properties of growth as being based on genetic evaluation of Korean native cattle. Analytic data were collected from 973 heads of male calves tested at Korea Native Cattle Improvement Centre from 1985 to 1988. Body weight at birth, weaning. yearling, and gain from weaning to yearling were affected highly(99.99%) by testing year-season and location at birth but the effects of parity was a little significant for growth traits. As data of chest girth at yearling showed highly skewed distribitions. it might be requested for transformation of data for genetic evaluation.

한우의 주요 경제형질에 대한 유전모수 추정 2 . Henderson 방법에 3 에 의한 증체형질들에 대한 유전력 및 유전상관 (遺傳相關)

이득환,오봉국,박영일,신언익,양영훈 ( D . H . Lee,B . K . Ohh,Y . I . Park,O . Y . Shin,Y . H . Yang ) 한국축산학회 1991 한국축산학회지 Vol.33 No.7

Heritabilities and genetic correlations were estimated by Henderson`s method 3 on the basis of the data from 867 heads of Korean native cattle tested at Korea Native Cattle Improvement Center. The traits for estimation of genetic parameters were body weights at birth, weaning, yearling. and 18-month of age, and chest girth at yearling age. The heritability estimates were 0.12±0.08 at birth weight, 0.31±0.17 at weaning weight. 0.11±0.08 at gain after weaning, and 0.19±0.07 at yearling weight. The genetic correlation between birth weight and gain after weaning was negatively estimated(-0.60), but that between gain after weaning and yearling weight was positively estimated(0.66).

축소개체모형을 이용한 한우의 증체형질에 대한 다형질 평가

이득환 ( D . H . Lee ) 한국축산학회 1992 한국축산학회지 Vol.34 No.1

Best linear unbiased predictors(BLUP) of breeding values for body weights at 18 months of age(18W) as single trait evaluation and those for 3 growth traits as multiple trait evaluation were estimated by indirect solving methods with reduced animal model(RAM) in Korean Native Cattle. Of 1629 animals reared at Korean Native Cattle Improvement Center, only 867 heads had valid records for all analytic traits. Breeding values for 18W could be estimated at 16 rounds of iteration by successive overrelaxation method with 1.4 relaxation factor and 2×10^-4 convergence criteria. In multiple trait evaluation involving birth weight(BW), weaning weight (WW) and postweaning gain(PG) with 2361 of mixed model equations, estimates by 1×10^-4 convergence criteria could be obtained at 29 rounds of iteration, when block iteration method with 1.2 relaxation factor was used, and also approximate prediction error variances could be estimated. The range of breeding values for WW was -37.5∼48.4㎏ and that for PG was -17.9∼16.6㎏ in multiple trait evaluation.

유전 및 육종 : 가축 유전체정보 활용 종축 유전능력 평가 연구 -표지인자 효과 추정 모의실험

이득환 ( Deuk Hwan Lee ),조충일 ( Chung Il Cho ) 한국동물자원과학회(구 한국축산학회) 2011 한국축산학회지 Vol.53 No.1

본 연구는 유전체분석에 대해 모의실험한 연구로써 Reference Population(RP)이 구성되었을 때, 표현형 자료가 없고 유전체자료만 있는 Juven1 또는 Juven2 세대에 대해 유전평가의 정확도에 대해 알아보고자 연구를 실시하였다. 모의실험의 가정으로 염색체는 1개이며 염색체길이는 100cM로 가정하였다. 초기의 유효집단의 수는 100두의 다형성이 없는 초기집단에서 유전자 효과가 없는 표지인자(Marker)를 0.1cM 및 0.5cM 간격으로 균등하게 단일 염기 돌연변이에 의한 다형성을 발생시켰고 유전자 효과가 있는 QTL 좌위는 Marker와 동수의 비율로 임의위치를 지정하여 돌연변이에 의한 변이성을 생성하였으며 이때 유전자 효과는 Gamma 분포함수(scale=1.66, shape=0.4)에서 생성하였다. 배우자(gamete) 형성과정에서 Haldane의 가정하에 유전자 재조합을 생성하였으며 돌연변이 발생율은 Marker 및 QTL 좌위에서 2.5×10(-3) 및 2.5×10(-5)의 확률로 발생시켜 1000세대까지 세대번식을 유지하였다. 이 후 1001세대부터 1004세대까지 세대당 2000두의 자손을 생성하였으며 이 때 유전력을 0.1 및 0.5의 가정하에 1001~1002세대에서 표현형 자료를 생성하였고, 1003~1004세대는 오직 유전체자료만 생성하였다. Bayesian 방법을 이용하여 개체별 육종가를 추정하였으며 표지인자간 거리(0.1cM, 0.5cM), 유전력(0.1, 0.5) 및 반형매 집단크기(20두, 4두)에 따라 참육종가와 추정 육종가간의 상관으로 표현되는 육종가 정확도에 대해 비교한 결과 1003세대에서 표지인자간 거리가 0.1cM 및 0.5cM일 때 육종가의 정확도는 각각 0.87, 0.81였고, 유전력이 0.1 및 0.5 일 때 각각 0.87, 0.94로 추정되었으며, 반형매 집단의 크기가 20두 일 때 0.87, 4두 일 때 0.84로 추정되었다. 위의 결과로 미루어 보아 다량의 SNP 표지정보 및 반형매 집단의 크기가 클수록 즉, 혈연계수가 높은 집단일 때 육종가의 정확도는 높게 나타났다. 유전체선발의 활용시 비교적 높은 정확도로써 조기선발이 가능하며 이로 인한 세대간격을 단축시킬 수 있어 개량의 효율을 높일 수 있을 것으로 사료된다. 반면에 유전체선발은 분석비용이 비싸며, 지속적인 유전체 선발시 특정유전자 선호로 인한 유전적 부동(Genetic Drift) 현상이 발생될 수 있기 때문에 지속적인 SNP 발굴에 대한 노력이 필요한(Meuwissen 2003) 단점이 있으나 한우 또는 젖소와 같은 대가축과 같이 세대간격이 긴 가축에서 유전체선발 할 경우 조기선발로 인한 세대간격 단축과 유전평가의 높은 정확도(0.8이상)로 인해 개량의 효율을 극대화 할 수 있을 것으로 사료된다. This simulation study was performed to investigate the accuracy of the estimated breeding value by using genomic information (GEBV) by way of Bayesian framework. Genomic information by way of single nucleotide polymorphism(SNP) from a chromosome with length of 100cM were simulated with different marker distance(0.1cM, 0.5cM), heritabilities(0.1, 0.5) and half sibs families(20 heads, 4 heads). For generating the simulated population in which animals were inferred to genomic polymorphism, we assumed that the number of quantitative trait loci(QTL) were equal with the number of no effect markers. The positions of markers and QTLs were located with even and scatter distances, respectively. The accuracies of estimated breeding values by way of indicating correlations between true and estimated breeding values were compared on several cases of marker distances, heritabilities and family sizes. The accuracies of breeding values on animals only having genomic information were 0.87 and 0.81 in marker distances of 0.1cM and 0.5cM, respectively. These accuracies were shown to be influenced by heritabilities (0.87 at h2=0.10, 0.94 at h2=0.50). According to half sibs`` family size, these accuracies were 0.87 and 0.84 in family size of 20 and 4, respectively. As half sibs family size is high, accuracy of breeding appeared high. Based on the results of this study it is concluded that the amount of marker information, heritability and family size would influence the accuracy of the estimated breeding values in genomic selection methodology for animal breeding.

한우의 종모우 평가방법에 관한 연구 2 . 선발지수법과 BLUP 에 의한 단형질 및 다형질 평가방법의 비교

이득환(D . H . Lee),이문걸(M . G . Lee),신언익(O . Y . Shin) 한국축산학회 1991 한국축산학회지 Vol.33 No.12

As a fundametal study of the sire evaluation method for simultaneous improvement of growth and carcass traits of Korean native Cattle (HANWOO). best linear unbiased predictors (BLUPs) by multitrait evaluation were compared with BLUPs by single trait evaluation and index values by selection index equation. Data on 510 male calves obtained from Korean Native Cattle Improvement Center were used for sire evaluation BLUPs for expected progeny differences in single trait evaluation with mixed model ranged -9.79∼12.91 ㎏ at body weight at 6 months of age(BW6), -31∼42 g/day at average daily gain(ADG), -0.87∼0.63% at retail cuts (RC) and -0.61∼0.85 score at marbling score(MS). The selection index equation estimated by BW6, ADG, MS was: BW6 + 85.18ADG + 4.93MS. BLUPs in multitrait evaluation were estimated wider ranges than those in single trait evaluation. In ,ire evaluation method, simulaneously considered BW6. ADG and MS. sire ranks by multitrait evaluation were positively correlated with those be selection index (0.94).

한우의 증체형질에 있어서 가기대법들에 의한 분산성분 추정

이득환 ( D . H . Lee ) 한국축산학회 1991 한국축산학회지 Vol.33 No.10

Three pseudoexpectation methods for variance component estimation were compared. Growth traits of Korean Native Cattle were analyzed for this study. Sire and error variances were estimated according to a mixed model that included fixed year-season, herd, parity and random sire and error effects. Quadratic forms estimating variance components and best linear unbiased prediction after absorbing all fixed effects into the equations for random effects were computed and variance components were estimated by pseudoexpectation methods. Hudson`s pseudoexpectation method was most quickly converged to estimates and the estimates obtained by Schaeffer`s pseudoexpectation method did not converge as quickly as the estimates for Van Raden`s method. The modified pseudoexpectation method of Van Raden had the highest correlation with restricted maximum likelihood estimates.

한우의 증체형질에 있어서 BLUP 에 의한 다형질 종모우 (種牡牛) 평가

이득환 ( D . H . Lee ) 한국축산학회 1992 한국축산학회지 Vol.34 No.1

Single- and multiple-trait sire evaluations were carried out on the basis of the data collected from 861 heads of male calves of Korean Native Cattle(HANWOO) tested at Korean Native Cattle Improvement Center from 1985 to 1988. Comparison was made between the best linear unbiased predictors with relationship and those without relationship of the single- and multiple-trait evaluation in sire mixed model. The range of the best linear unbiased predictors without relationship was wider than that with relationship in single-trait evaluation and predictors of sire`s genetic effects in multiple-trait evaluation with relationship showed the larger variety than those in single-trait evaluation with relationship. The range of the best linear unbiased predictors for weaning weight was -5.30∼10.28㎏ and that for postweaning gain was -5.05∼ 2.61㎏ in multiple-trait sire mixed model with relationship.

유전 및 육종 : 특정 종돈집단의 임신기간, 이유후초종부일, 총산자수 및 사산에 대한유전모수 추정

이득환 ( Deuk Hwan Lee ),손지현 ( Ji Hyun Son ) 한국동물자원과학회(구 한국축산학회) 2013 한국축산학회지 Vol.55 No.5

Estimate breeding value can be used as single trait animal model was developed directly using the Fortran language program. The program is based on data computed by using the indirect method repeatedly. The program develops a common algorithm and imprves efficiency. Algorithm efficiency was compared between the two programs. Estimated using the solution is easy to farm and brand the service, pedigree data base was associated with the development of an improved system. The existing program that uses the single trait animal model and the comparative analysis of efficiency is weak because the estimation of the solution and the conventional algorithm programmed through regular formulation involve many repetition; therefore, the newly developed algorithm was conducted to improve speed by reducing the repetition. Single trait animal model was used to analyze Gauss-Seidel iteration method, and the aforesaid two algorithms were compared thorough the mixed model equation which is used the most commonly in estimating the current breeding value by applying the procedures such as the preparation of information necessary for modelling, removal of duplicative data, verifying the parent information of based population in the pedigree data, and assigning sequential numbers, etc. The existing conventional algorithm is the method for reading and recording the data by utilizing the successive repetitive sentences, while new algorithm is the method for directly generating the left hand side for estimation based on effect. Two programs were developed to ensure the accurate evaluation. BLUPF90 and MTDFREML were compared using the estimated solution. In relation to the pearson and spearman correlation, the estimated breeding value correlation coefficients were highest among all traits over 99.5%. Depending on the breeding value of the high correlation in Model Ⅰ and Model Ⅱ, accurate evaluation can be found. The number of iteration to convergence was 2,568 in Model Ⅰ and 1,038 in Model Ⅱ. The speed of solving was 256.008 seconds in Model Ⅰ and 235.729 seconds in Model Ⅱ. Model Ⅱ had a speed of approximately 10% more than Model Ⅰ. Therefore, it is considered to be much more effective to analyze large data through the improved algorithm than the existing method. If the corresponding program is systemized and utilized for the consulting of farm and industrial services, it would make contribution to the early selection of individual, shorten the generation, and cultivation of superior groups, and help develop the Hanwoo industry further through the improvement of breeding value based enhancement, ultimately paving the way for the country to evolve into an advanced livestock country.

축소개체모형에서 가기대법을 이용한 한우의 증체형질에 대한 분산성분 추정

이득환 ( D . H . Lee ) 한국축산학회 1991 한국축산학회지 Vol.33 No.10

Variance components for growth traits of Korean Native Cattle were estimated by a pseudoexpectation method in the reduced animal model. This method was quickly converged to estimates of variance components for growth traits of Korean Native Cattle. There were a little differences between estimates of variance com ponents by the pseudoexpectation method in the reduced animal model and those by the restricted maximum likelihood in the sire mixed model.