Soybean (Glycine max (L.) Merr.) is an important legume crop with high commercial value widely cultivated globally. Thus, the genetic characterization of the existing soybean germplasm will provide useful information for enhanced conservation, improvement and future utilization. This study aimed to assess the extent of genetic diversity of soybean elite breeding lines and varieties developed by the soybean breeding programme of the International Institute of Tropical Agriculture (IITA), Ibadan, Nigeria. The genetic diversity of 65 soybean genotypes was studied using single-nucleotide polymorphism (SNP) markers. The result revealed that 2446 alleles were detected, and the indicators for allelic richness and diversity had good differentiating power in assessing the diversity of the genotypes. The three complementary approaches used in the study grouped the germplasm into three major clusters based on genetic relatedness. The analysis of molecular variance revealed that 71% (P < 0.001) variation was due to among individual genotypes, while 11% (P < 0.001) was ascribed to differences among the three clusters, and the fixation index (FST) was 0.11 for the SNP loci, signifying moderate genetic differentiation among the genotypes. The identified private alleles indicate that the soybean germplasm contains diverse variability that is yet to be exploited. The SNP markers revealed high diversity in the studied germplasm and found to be efficient for assessing genetic diversity in the crop. These results provide valuable information that might be utilized for assessing the genetic variability of soybean and other legume crops germplasm by breeding programmes.

In farmed fish, selective breeding for feed conversion ratio (FCR) may be possible via indirectly selecting for easily-measured indicator traits correlated with FCR. We tested the hypothesis that rainbow trout with low lipid% have genetically better FCR, and that lipid% may be genetically related to retention efficiency of macronutrients, making lipid% a useful indicator trait. A quantitative genetic analysis was used to quantify the benefit of replacing feed intake in a selection index with one of three lipid traits: body lipid%, muscle lipid% or viscera% weight of total body weight (reflecting visceral lipid). The index theory calculations showed that simultaneous selection for weight gain and against feed intake (direct selection to improve FCR) increased the expected genetic response in FCR by 1·50-fold compared with the sole selection for growth. Replacing feed intake in the selection index with body lipid%, muscle lipid% or viscera% increased genetic response in FCR by 1·29-, 1·49- and 1·02-fold, respectively, compared with the sole selection for growth. Consequently, indirect selection for weight gain and against muscle lipid% was almost as effective as direct selection for FCR. Fish with genetically low body and muscle lipid% were more efficient in turning ingested protein into protein weight gain. Both physiological and genetic mechanisms promote the hypothesis that low-lipid% fish are more efficient. These results highlight that in breeding programmes of rainbow trout, control of lipid deposition improves not only FCR but also protein-retention efficiency. This improves resource efficiency of aquaculture and reduces nutrient load to the environment.

Genetic improvement of farm animals encompasses both mating and selection for desired traits and indigenous chicken genetic resources are no exception. In Kenya, previous attempts to genetically improve indigenous chicken involved cross-breeding scheme by utilizing cockerels and pullets of exotic breeds with the local indigenous chicken. This scheme was complimented with farmer training on good management practices and vaccination for disease control. The scheme was partially successful with improved performance in the crossbreds that declined with subsequent generations. Failure of the programme to meet stakeholder's expectation led to its’ termination. The current attempt through the Smallholder Indigenous Chicken Improvement Programme has initiated an holistic and comprehensive approach to analysing the entire indigenous chicken actors and avert the causes of previous failures. The programme has genetically and phenotypically characterized the chicken; established reference/base population collected from different ecotypes/counties, established the breeding goals and designed breeding programmes that best suit the Kenyan stakeholders. The on-station research has reported variation on production traits, determined heritability estimate on growth. Current and ongoing research is focused on molecular characterization, selection for improved immune response, carcass quality, eggs production, growth and adaptation traits. The research is also concerned with conservation of these genetic resources.

A dynamic selection algorithm for maximizing annual genetic response while constraining the rate of inbreeding per generation in populations with overlapping generations is presented. The procedure gives the optimum number of individuals to be selected and the progeny they each produce. The solution to the problem was obtained by using BLUP estimated breeding values, the augmented numerator relationship matrix and lifetime breeding profiles. The procedure was able to constrain the rate of inbreeding per generation to a predefined level across generations of selection by considering all gene flow pathways. The optimization procedure represents an improvement on standard truncation BLUP selection, as it yielded substantially more genetic response (up to 35%) at the same rate of inbreeding.

Methods that maximize genetic response in populations with overlapping generations while controlling rate of inbreeding by constraining the average relationship among selection candidates were compared. Firstly, computer simulations of closed nucleus selection schemes showed that a two-stage optimization algorithm approach, where the distribution of parents within and thereafter over age classes was optimized resulted in different breeding schemes than an approach that performed an iteration on this distribution. It yielded significantly lower annual genetic gain (0·194 v. 0·223 σp units), fewer animals selected (21·9 v. 26·4) and longer generation intervals (2·38 v. 1·68 years) but maintained the rate of inbreeding closer to its constraint. In large schemes, iteration may be computationally the only feasible method for the optimization of parents across age classes. Secondly, the use of conventional relationships for constraining inbreeding was compared with that of augmented relationships, which do not depend on the level of inbreeding. Both relationships resulted in very similar breeding schemes, but the use of augmented relationships avoids correction of the current level of inbreeding. Thirdly, a constraint of the rate of inbreeding on a per year basis was compared with a constraint on a per generation basis. When optimizing per generation, the generation interval was shorter compared with a scheme where an analogous annual restriction was in place (2·01 v. 2·38 years) and the annual rate of genetic gain was higher (0·214 v. 0·194 σp units).

The effects on genetic and economic responses of adding faecal egg count (FEC) in the aggregate genotype of dairy sheep in Greece were investigated. The extra responses obtained in the full aggregate genotype were expressed as a percentage change of the responses in terms of genetic gain, profit and selection response of single traits before adding the trait. The initial aggregate genotype included the traits milk fat yield (FY) and number of lambs weaned per ewe per year (NLW). Inclusion of FEC in the aggregate genotype resulted always in increased genetic gain and profit. The extra responses from adding the FEC in the index selection were variable and often very large, depending on the parameters varied i.e. the economic weight of FEC and the genetic correlation of FEC with the other production traits. For a wide range in the size of the genetic correlations between FEC and other traits, gains of FEC and no appreciable losses of responses for FY and NLW were predicted when FEC accounted proportionately for 0·10 to 0·20 of the total monetary genetic deviation. While FEC showed a wide range of change, selection responses of FY and NLW remained remarkably insensitive under various weightings of FEC and different genetic correlations of FEC with the production traits. The genetic and economic efficiency of alternative selection schemes were also evaluated. A two-stage selection procedure involving preliminary selection of rams on dam’s FEC, FY, NLW and on their own FEC and final culling on progeny’s FEC, FY and NLW was predicted to be the most efficient in both genetic and economic terms. Female replacements should be selected on dam’s FEC, FY and NLW (first stage) and their own FEC, FY and NLW (second stage). When repeated measurements of FEC are taken, the recommended number of FEC measurements was found to be 4.

A model which was originally developed to derive the probability density of a calving distribution for beef cows under natural breeding has been modified to incorporate the effects of using oestrous synchronization at the beginning of the breeding period.

The model has been used to investigate five oestrous synchronization strategies over four herds, each with differing calving patterns: no synchronization; every cow in the herd is synchronized; every anoestrous cow is synchronized; every cow calving more than 20 days after the date of the planned start of calving is synchronized; every cow calving 30 days after the date of the planned start of calving is synchronized. Two of the herds previously had long calving periods and two previously had short, compact calving periods.

Model results showed that synchronization improved average herd calf live weight by 21 kg in the two herds with long prior calving periods but showed little effect in the two herds with compact prior calving periods. Synchronizing cows which calved more than 20 days after the start of calving in the herd showed similar results to synchronizing every cow in the herd and thus would be more cost effective. Synchronizing cows which calved more than 30 days after the start of calving in the herd showed a smaller advantage but only in the herds with long prior calving periods. The optimal strategy for oestrous synchronization in a herd will depend on the prior calving distribution and the average duration ofanoestrus in that herd as well as the relative costs of the treatment.

Many decades of selection for increased egg output per bird and for efficient feed conversion have resulted in stocks capable of remarkable levels of performance. In the search for new ways to achieve further improvements in egg production efficiency, residual feed intake is seen as a parameter that merits the serious attention of breeders. This review examines relevant research relating to feed conversion efficiency since 1972 – including some results obtained by the author from commercial populations – and offers guidelines on the ways in which datd on individual feed intake could be applied within modern selection programmes.

Predicted transmitting abilities for somatic cell counts (SCC) are available in the United Kingdom and there is a direct economic benefit attached to reducing SCC as the milk payment schemes include a penalty for high SCC levels in bulk tank samples and sometimes a premium for low SCC. The aim of the present study was to establish the economic importance of bulls' breeding values for SCC in relation to this payment scheme for SCC. To do this, an empirical method was developed using 645071 individual cow SCC and milk yield test-day records from 358 herds. The economic value was calculated by (i) decreasing all individual cow records by 0·01 and comparing the average penalty with the current average penalty, and (ii) taking the derivative of a Gompertz function describing the within-herd penalty per cow as a function of the average within-herd 3-month rolling geometric mean SCC. Mean milk test-day yield and average test-day SCC were 20·4 kg and 262 kcount per ml respectively with, on average, 83 cows tested each day. In the current situation the average penalty paid was 0·54, 0·18 and 3·2 pence per litre depending on whether an England and Wales, Scottish or future payment scheme was used, respectively. Across the population, the economic values per 0·01 decrease in SCC were £1·04, £0·54 and £6·03 per cow per year for these three payments schemes respectively. However these economic values depend strongly on the mean SCC. Herds have different means and as the future population mean is difficult to predict, it is suggested that for herds with the majority of their bulk tank samples in penalty bands 1, 2 or 3+ (average SCC of <150, 150 to 250 and 250+) the economic values are £0, £0·50, and £15 per cow per year per 0·01 reduction in SCC, respectively, until additional benefits have been quantified.

Incorporation of food intake into breeding objectives for ruminants is often difficult because the relevant genetic parameters are unavailable. A non-statistical approach to estimating the genetic relationships between food costs and genetic changes in growth and body composition traits was therefore developed.

Equations which predict additional food requirements associated with genetic changes in growth and body composition traits are derived from an interspecies growth model. In the growth model, a genotype is mainly described according to its expected mature weights of protein, lipid and a composition trait of interest in the empty body under non-limiting conditions. Other model parameters describe the expected growth of the animal in a specific environment.

An example is shown for beef cattle in the United Kingdom, where increases in carcass weight by 1 kg and carcass subcutaneous fat proportion by 0·01 result in increases in food costs by £0·97 and £7·34 respectively for a production system where animals are given an intensive silage/concentrate diet, are slaughtered at 305 kg and have a commercial fat grade 4L. A strong dependence of the results on the type of production system, especially the degree of maturity in protein at slaughter, is demonstrated.

The model is easily adapted to other breeds and ruminant species. The approach is recommended for incorporating food intake into breeding objectives for ruminants in situations where reliable estimates of the necessary genetic parameters for a more conventional specification of the breeding objective are unavailable.

Optimum designs of three-tier open nucleus breeding schemes are evaluated deterministically by maximizing the equilibrium rate of genetic gain for two methods of selection. Methodology is developed for both restricted and unrestricted migration between tiers and incorporates the loss of variance due to selection. A formula is derived for calculating the asymptotic rate of inbreeding. In the extensive livestock industries, proportional improvements in the equilibrium rate of genetic gain of between 0·12 and 0·22 are possible over a closed nucleus if no restrictions are imposed on male and female migration between tiers. The value of the extra tier in an optimized three-tier open nucleus scheme is approximately 0-45 of the maximum proportional improvement of a two-tier open nucleus over a closed nucleus scheme. The optimum structure is to have approximately 1% and 10% of the population in the nucleus and multiplier respectively. With this optimum structure the asymptotic rate of inbreeding will be reduced to one-sixth of that in a closed nucleus. The effects of various factors on the optimum structure, genetic gain and inbreeding are examined. The advantage of a three-tier open nucleus scheme over a closed nucleus scheme can be reduced substantially if thesefactors are operating.

The genetic and economic efficiency of alternative selection schemes and breeding objectives for the Karagouniki dairy sheep in Greece was investigated by model calculations. Criteria of efficiency were the annual genetic gain for the aggregate breeding value, the profit per ewe in the population and the annual selection responses for single traits. The introduction of a two-tier selection scheme, where the recorded ewes are separated into a nucleus and a pre-nucleus, was found superior in both genetic and economic terms for a breeding objective comprising the traits milk yield and number of lambs weaned per ewe per annum. Highest rates for annual genetic gain and selection responses were obtained when the nucleus size was proportionately 0·05 of the population (200 000 ewes), the size of the test matings 0·50 to 0·60 of the size of the pre-nucleus unit (0·10 of the population) and the number of daughters per test ram 40. Opening the nucleus to replacement ewes from the lower tier did not affect positively the annual genetic gain and the selection responses. Furthermore, a breeding objective comprising the traits fat yield and number of lambs weaned per ewe and per annum was found very efficient in both genetic and economic terms while selection on growth and carcass traits did not seem to be justified.

A selection programme was set up in 1983 in the Ivory Coast to improve the growth and live weight of the indigenous Djallonkè sheep using an open nucleus breeding system. Selection was based on male individual weights at 80, 180 and 365 days of age. Multiple sires were used in farmers' flocks. Individual animal model, using average numerator relationship was used to estimate breeding values from which genetic trends were derived. This paper reports results of the analyses of 10 417 records of 80-day weights (WT80) of lambs born between 1984 and 1992 from 29 participating farmers, and 1978 and 849 records on 180- (WT180) and 365- (WT365) day weights, respectively, of lambs from the nucleus. Phenotypic trends exhibited substantial annual fluctuation during the study period. The heaviest weights were obtained in 1986 (10·1 kg) for WT80, 1984 (23·3 kg) for WT180 and 1985 (33·7 kg) for WT365. There was an annual decline of 152 g (P<0·05) in WT80 mainly (–180 g/year) due to negative environmental trend. The phenotypic and environmental trends for WT180 were –703 and –721 g/year, respectively, while the corresponding values for WT365 were –931 and –956 g/year. Overall mean weights of the ram lambs at the three ages were 9·9, 23·3 and 33·7 kg, respectively in the base year and 9·3, 17·8 and 28·2 kg, respectively in the final year. Breeding values increased by 28, 11 and 14 g/year for WT80, WT180 and WT365, respectively. Results of this study indicate that genetic progress can be made in growth performance of Djallonkè sheep if reasonable levels of animal management as well as selection pressure are maintained.

Data on Djallonke sheep collected from 1983 through 1992 on 29 participating farms of an open nucleus improvement programme and from the nucleus were analysed. Initial on-farm records on 14342 lambs and on-station records on 2551 ram lambs were edited down to 6605 for birth weight (BWT), 5857 for preselection (birth to 80 days) average daily gain (PSADG), 10417 for 80-day weight (WT80), 13054 for lamb survival at pre-selection, 1978 for 180 (WT180) and 849 for 365 (WT365) day weights. Year of birth, sire line, flock, month and season of birth had significant (at leastP < 0·05) effects on both on-farm and on-station weights and on lamb survival. Trait means were 2·2 (s.e. 0·02) kg for BWT, 69·6 (s.e. 1·26) g/day for PSADG, 9·1 (s.e. 0·05) kg for WT80, 19·7 (s.e. 0·24) kg for WT180 and 31·8 (s.e. 0·45) kg for WT365. Mean survival was 90 (s.e. 0·9) %. Single lambs were heavier at birth and at 80 days of age, grew faster to 80 days and were about 200 and 500 g heavier at 180 and 365 days respectively than multiple lambs. There was a large variation between flocks: flock means for WT80 varied from 6·4 to 12·0 kg (CV 0·21), with smaller flocks having generally the lowest means. Lambs born in the hot-rainy season had the heaviest birth and 80-day weights, whereas ram lambs born in the cool rainy and early hot-dry seasons had the heaviest 180- and 365-day weights. Lambs born during the cool months of July to October had the lowest survival rate. Multiplicative factors were found to be more appropriate for adjusting on-farm records for type of birth, management level, birth date, season of birth and ewe parity than additive adjustment factors.

A procedure for maximizing genetic gain (after a number of generations of selection) for a given rate of inbreeding or for a given coefficient of variation of response is presented. An infinitesimal genetic model is assumed. Mass selection is practised for a number of discrete generations. With constraints on inbreeding, expected rates of genetic progress (ΔG) are combined with expeced rates of inbreeding (ΔF) in a linear objective function (Ω = ΔG - μΔF). In addition, an expression to approximate the rate of gain at any generation accounting for changes in genetic parameters due to linkage disequilibrium and due to inbreeding is derived. Predicted gain is in general within 5% of that obtained from simulation. Thus, both ΔG and ΔF are obtained from simple analytical formulae. An equivalent function is used when the coefficient of variation of response (CV) is the parameter restricted (Ω = ΔG -μCV). Maximization of the objective function Ω for appropriate values of μ gives the optimum number of sires and dams selected when specific constraints on the level of inbreeding or the coefficient of variation of response are imposed. The method is applied to a practical situation in fish breeding. Optimum mating ratios and optimum numbers of sires selected are obtained for different scored population sizes and heritabilities. Results obtained with this procedure agree very well with results from simulation studies. The optimum number of sires increases with the size of the scheme and with more severe restrictions on risk. In the schemes considered, the optimum mating ratio is equal to 2 unless the constraint on the rate of inbreeding is severe, the size of the scheme is small and the heritability is low. In these situations the optimum mating ratio is equal to 1. The procedure is general in terms of generations of selection considered and in terms of parameters to be constrained. A large amount of computer processor unit time is saved with this method in comparison with simulation procedures.

Customized sire selection indices were developed for Australian dairy bulls in order to promote more objective use of estimated breeding values in commercial breeding programmes. It was assumed that the breeding goal for commercial dairy farmers is the profitability of a bull's progeny. Seven characteristics of the progeny were identified as having a major impact on profitability and were included in the breeding objective: milk, fat and protein yield, survival, body weight, milking speed and temperament. Traits in the selection indices used to predict profit were milk, fat and protein yield, survival, milking speed, temperament, size, overall type and fore teat placement. Size was included because of its correlation with body weight, and overall type and front teat placement because of their correlation with survival. To avoid double counting the benefits of milk production traits, temperament and milking speed, the survival trait in the objective was defined as survival independent of voluntary culling for these traits. Customization of the breeding objective was achieved by adjusting the economic weights for traits in the objective to take account of important characteristics of farmers' herds, the milk payment system under which they operate and make allowance for their own value judgements. An assessment of the impact of customization suggested that, even though there is a wide range in the economic weights that are applicable in different areas of Australia, there would be little loss of efficiency in using a single national index. However, customization is still believed to be desirable given that it is likely that a substantial proportion offarmers will be reluctant to use a national index, especially in those states which have quotas and focus on the liquid milk market. The algorithms described in the paper have been incorporated into a user-friendly microcomputer program called $electabull which is now commercially available to farmers.

Commercial application of multiple ovulation and embryo transfer (MOET) technology will be subject to practical constraints and economic rationalism. This study examines use of MOET in its most profitable arena: to breed stud rams which will disseminate genetic improvement widely through multiplier studs to commercial flocks. A deterministic prediction is used to evaluate schemes based on an open nucleus MOET group within a Merino parent stud, taking account of genetic merit and inbreeding. Selection is based on clean fleece weight with an assumed heritability of 0·4. Embryos are collected at a rate equivalent to 3·45 live lambs per donor. Benefits of MOET were calculated from the discounted expressions of rams sold, and compared with the costs incurred.

As the proportion of the flock born from MOET increases, the rate of genetic gain increases rapidly at first, but diminishing returns are observed. The costs ofMOET increase linearly with the number of lambs produced, so the optimum proportion ofMOET lambs is for practical purposes always less than 100%.

Some use of MOET was profitable provided the stud sells sufficient stud rams each year. Sensitivity tests found that other parameters had only a small impact on the optimum level ofMOET. In general however, changes which increased the rate of genetic gain (heritability, flock size) or increased its value (wool price, lower discount rate) increased the optimum number ofMOET lambs.

The results should provide guidelines to optimum investment in MOET for the wool industry. An across flock genetic evaluation scheme is probably necessary to motivate this investment.

Stochastic simulation was used to compare selection response and rate of inbreeding in four nucleus breeding systems and a sire referencing scheme for sheep: an open nucleus system (ONS), an open nucleus system with sequential selection of the nucleus (ONS-S), a sire referencing scheme (SRS) and a dispersed open nucleus system (DONS). Selection was based on best linear unbiased prediction of breeding values for a single trait measurable on all individuals prior to selection. Selection in a population of 1200 ewes equally divided into 10 flocks was simulated over a 15-year period. The mean rate of genetic gain was proportionately about 0-15 higher in ONS-S and DONS compared with ONS and SRS. The rate of inbreeding in SRS was considerably lower and in ONS-S, considerably higher, than in the other systems. The level of prolificacy in the population did not influence the relative ranking of the breeding systems but may have implications for their optimal structure.

Alternative closed breeding schemes for beef cattle are analysed using stochastic computer simulation. Multiple ovulation and embryo transfer (MOET) schemes are compared with conventional schemes (schemes without MOET) with an equal expected number of progeny born per year. Schemes are compared for genetic gain and inbreeding obtained after 25 years of selection. The trait considered, evaluated in both sexes, has an initial heritability of 0·35. Different population sizes and numbers of sires selected are evaluated. Current realistic parameters for embryo production are assumed in MOET schemes.

After 25 years of selection, and with no control on inbreeding, cumulative genetic gains are about 50% higher in MOET schemes compared with conventional schemes. The benefit from MOET is mostly due to increased selection intensities in females. The rate of inbreeding increases by up to nearly 300% when MOET is used. This maximum percentage increase in inbreeding following the use of MOET can be reduced to about 100% when selection and mating strategies for controlling inbreeding are used. The effect of the number of sires used on the inbreeding obtained is more important than the effect of the size of the herd. In MOET schemes, increasing the number of sires selected by a factor of three, leads to reductions in inbreeding rates of 40%. When schemes of the same size are compared at similar acceptable inbreeding levels, MOET schemes give around 30% higher genetic progress than conventional schemes.

Stochastic simulation was used to compare the results of alternative breeding systems in a sheep population divided into 10 flocks of 120 ewes. The breeding systems compared were selection within closed flocks (CF), a closed nucleus system (CNS), an open nucleus system (ONS) and open nucleus systems with the selection of nucleus replacements being restricted to either nucleus born males (ONSRm) or nucleus born females (ONSRf). Selection was for a best linear unbiased prediction of breeding value for lamb live weight which had a heritability of 0·17. The open nucleus breeding systems (ONS, ONSRm, ONSRf) resulted in higher rates of genetic gain, more predictable selection responses and lower rates of inbreeding than either the closed nucleus system (CNS) or selection within closed flocks (CF). Initial genetic differences between flocks resulted in higher rates of genetic gain in the nucleus breeding systems due to the use of between flock genetic variance. In the ONS system up to 25% of nucleus sires and approximately 50% of nucleus dams were born in base flocks. Nevertheless if selection of either nucleus sires or dams was restricted to nucleus born animals there was very little change in genetic gain or rate of inbreeding.