Douglas Scott Falconer, or D. S. Falconer as he normally styled himself, died on 23 February 2004, within days of his 91st birthday. His health had deteriorated and his eyesight, sadly, was failing, but he was able to retain many of his interests until his brief final illness. He had been for many years a prominent member of a remarkable group of geneticists that the late C. H. Waddington had gathered in Edinburgh after the second world war, and for a long time he was regarded as a leading authority in the area of quantitative genetics.

In Drosophila melanogaster, the hobo transposable element is responsible for a hybrid dysgenesis syndrome. It appears in the germline of progenies from crosses between females devoid of hobo elements (E) and males bearing active hobo elements (H). In the HE system, permissivity is the ability of females to permit hobo activity in their progeny when they have been crossed with H males. Permissivity displays both intra- and inter-strain variability and decreases with the age of the females. Such characteristics are reminiscent of those for the reactivity in the IR system. The reactivity is the ability of R females (devoid of I factors) to permit activity of the I LINE retrotransposon in the F1 females resulting from crosses with I males (bearing I factors). Here we investigated permissivity properties in the HE system related to reactivity in the IR system. Previously it had been shown that reactivity increases with the number of Su(var)3-9 genes, which increases chromatin compaction near heterochromatin. Using the same lines, we show that permissivity increases with the number of Su(var)3-9 genes. To investigate the impact of chromatin compaction on permissivity we have tested the polymorphism of position-effect variegation (PEV) on the whitemottled4 locus in RE strains. Our results suggest a model of regulation in which permissivity could depend on the chromatin state and on the hobo vestigial sequences.

Population genetic analyses traditionally focus on the frequencies of alleles or genotypes in ‘populations’ that are delimited a priori. However, there are potential drawbacks of amalgamating genetic data into such composite attributes of assemblages of specimens: genetic information on individual specimens is lost or submerged as an inherent part of the analysis. A potential also exists for circular reasoning when a population's initial identification and subsequent genetic characterization are coupled. In principle, these problems are circumvented by some newer methods of population identification and individual assignment based on statistical clustering of specimen genotypes. Here we evaluate a recent method in this genre – Bayesian clustering – using four genotypic data sets involving different types of molecular markers in non-model organisms from nature. As expected, measures of population genetic structure (FST and ΦST) tended to be significantly greater in Bayesian a posteriori data treatments than in analyses where populations were delimited a priori. In the four biological contexts examined, which involved both geographic population structures and hybrid zones, Bayesian clustering was able to recover differentiated populations, and Bayesian assignments were able to identify likely population sources of specific individuals.

Introgression of Drosophila simulans genes into the Drosophila melanogaster genome provides an ideal system for analysing genetic incompatibility between species. Females and males homozygous for the introgression Int(2L)S (cytologically, 30F3-31C5 to 36A2-7) are sterile. Genetic dissection of the proximal part of the introgression (34D1-3 to 36A2-7) has indicated that introgressions of 0·7–1·6 Mb size result in male sterility when homozygous. In the present analysis we examine the distal part of the introgression (30F3-31C to 34D1-3) and reveal that introgressions with similar DNA content (1·8–2·1 Mb) result in male sterility. Compared with introgressions between the more closely related species Drosophila mauritiana and D. simulans, the minimal introgression resulting in male sterility is smaller by several-fold.

As an alternative to multiple-interval mapping a two-step moment method was recently proposed to map linked multiple quantitative trait loci (QTLs). The advantage of this moment method was supposed to be its simplicity and computational efficiency, especially in detecting closely linked QTLs within a marker bracket, but also in mapping QTLs in different marker intervals. Using simulations it is shown that the two-step moment method may give poor results compared with multiple-interval mapping, irrespective of whether the QTLs are in the same or in different marker intervals, especially if linked QTLs are in repulsion. The criteria of comparison are number of identified QTLs, likelihood ratio test statistics, means and empirical standard errors of the QTL position and QTL effects estimates, and the accuracy of the residual variance estimates. Further, the joint conditional probabilities of QTL genotypes for two putative QTLs within a marker interval were derived and compared with the unmodified approach ignoring the non-independence of the conditional probabilities.

Recent research in bovine genetics has focused on characterization of the biological differences underlying phenotypic variation for qualitative and quantitative traits of economic value in existing cattle populations. The much-anticipated benefits of DNA-based tools to routinely guide selection decisions for more efficient genetic gain and widened profit margins have not been fully met since the origin of this premise over two decades ago. However, the impending release of a high-quality draft genome sequence in 2005 should mark a turning point in these efforts. The following discussion summarizes how the bovine genetics research community has positioned itself to fully utilize a genome sequence resource and frames how genome sequence information can not only be applied to better implement marker-assisted selection, but also address rising consumer concerns relative to animal well-being and food safety.

We performed genetic analysis of hybrid sterility and of one morphological difference (sex-comb tooth number) on D. yakuba and D. santomea, the former species widespread in Africa and the latter endemic to the oceanic island of São Tomé, on which there is a hybrid zone. The sterility of hybrid males is due to at least three genes on the X chromosome and at least one on the Y, with the cytoplasm and large sections of the autosomes having no effect. F1 hybrid females carrying two X chromosomes from either species are perfectly fertile despite their genetic similarity to completely sterile F1 hybrid males. This implies that the appearance of Haldane's rule in this cross is at least partially due to the faster accumulation of genes causing male than female sterility. The larger effects of the X and Y chromosomes than of the autosomes, however, also suggest that the genes causing male sterility are recessive in hybrids. Some female sterility is also seen in interspecific crosses, but this does not occur between all strains. This is seen in pure-species females inseminated by heterospecific males (probably reflecting incompatibility between the sperm of one species and the female reproductive tract of the other) as well as in inseminated F1 and backcross females, probably reflecting genetically based incompatibilities in hybrids that affect the reproductive system. The latter ‘innate’ sterility appears to involve deleterious interactions between D. santomea chromosomes and D. yakuba cytoplasm. The difference in male sex-comb tooth number appears to involve fairly large effects of the X chromosome. We discuss the striking evolutionary parallels in the genetic basis of sterility, in the nature of sexual isolation, and in morphological differences between the D. santomea/D. yakuba divergence and two other speciation events in the D. melanogaster subgroup involving island colonization.

t haplotypes are naturally occurring, variant forms of the t complex on mouse chromosome 17, characterized by the presence of four inversions with respect to wild-type. They harbour mutations causing male sterility, male transmission ratio distortion (TRD) and embryonic lethality. Mice carrying t haplotypes have been found throughout the world, and genetic studies of the lethal mutations have identified at least 16 complementation groups. The embryonic lethal phenotypes of many t haplotypes have been characterized in detail, and are thought to be the consequence of homozygosity for single gene mutations. However, the existence of additional mutations in genes that function at later stages of development would be obscured. Here we investigated the possibility of multiple mutations in t haplotypes by screening the tw73 haplotype for the presence of novel mutations. Since genetic analysis of t haplotype mutations is hindered by recombination suppression due to the inversions, deletion complexes covering the proximal two-thirds of the t complex were used to uncover the presence of any new lethal alleles. This analysis revealed a novel mutation between D17Jcs41 and D17Mit100, causing mice carrying both tw73 and selected deletions to die at birth, prior to feeding. The finding of a new, cryptic lethal mutation in t haplotypes is an indication that these recombinationally isolated chromosomes, which already contain at least one lethal mutation that prevents homozygosity, may serve as sinks for the accumulation of additional recessive mutations.

Mating plans are investigated in order to minimize inbreeding in small populations when frozen semen is available. For a single dam line it was found that specific sire rotations minimized the asymptotic level of inbreeding when semen is used repeatedly from certain generations. When semen of N foundation (G0) sires is used rotationally over generations it is shown that the inbreeding level asymptotes to 1/(2N+1−2). However, if only G0 sires are used then all genes will eventually descend from the founder sires. Inbreeding can be reduced further by using sires from generation one (G1) and later as this retains genes from the founder dams in the long-term gene pool. If semen from NG0 sires and N unrelated G1 sons is used rotationally then inbreeding asymptotes to (2N−1+1)/(22N+1−2). When there are more founder dams than sires, the asymptotic inbreeding can be reduced even further by using the semen of half-sib G1 sires in rotation. Optimal rotations using full-sib G1 sires or generation 2 (or later) sires will lower the asymptotic inbreeding also, but generally not by much. It was found that when unlimited frozen semen from a specified group of sires was available, the optimal mating plan was achieved by selecting each generation the sire with the least co-ancestry with the current female of the dam line.

A paternally expressed QTL for muscle growth and backfat thickness (BFT) has previously been identified near the IGF2 locus on the distal tip of pig chromosome 2 (SSC2p) in three experimental F2 populations. Recently, a mutation in a regulatory element of the IGF2 gene was identified as the quantitative trait nucleotide (QTN) underlying the major QTL effect on muscle growth and BFT in crosses between Large White and Wild Boar or Pietrain. This study demonstrates that the IGF2 mutation also controls the paternally expressed QTL for backfat thickness in a cross between Meishan and European Whites. In addition, a comparison of QTL of backfat thickness measured by Hennessy grading probe (HGP) and by ultrasound measurement (USM) was made. In the USM analyses, the IFG2 mutation explains the entire QTL effect on SSC2p, whereas in the HGP analysis the presence of a second minor QTL can not be excluded. Finally, this study shows that this particular IGF2 mutation does not cause the paternally expressed QTL for teat number mapping to the same region of SSC2p as the BFT QTL.

Myostatin is a negative regulator of muscle growth and mutations in its gene lead to muscular hypertrophy and reduced fat. In cattle, this is seen in ‘double muscled’ breeds. We have used marker-assisted introgression to introduce a murine myostatin mutation, MstnCmpt-dl1Abc [Compact (C)], into an inbred line of mice (DUHi) that had been selected on body weight and had exceptional growth. Compared with homozygous wild-type mice, homozygous (C/C) mice of this line were ~4–5% lighter, had ~7–8% shorter tails, substantially increased muscle weights (e.g. quadriceps muscle in males was 59% heavier) and an increased ‘dressing percentage’ (~49% vs 39%), an indicator of overall muscularity. The weights of several organs (e.g. liver, kidney, heart and digestive tract) were significantly reduced, by 12–20%. Myostatin deficiency also resulted in drastic reductions of total body fat and of various fat depots, total body fat proportion falling from ~17·5% in wild-type animals of both sexes to 9·5% and 11·6% in homozygous (C/C) females and males, respectively. Males with a deficiency in myostatin had higher gains in muscle traits than females. Additionally, there was a strong distortion of the segregation ratio on the DUHi background. Of 838 genotyped pups from inter se matings 29%, 63% and 8% were homozygous wild type (+/+), heterozygous (C/+) and homozygous (C/C), respectively, showing that MstnCmpt-dl1Abc has lower fitness on this background. This line, when congenic, will be a useful resource in gene expression studies and for finding modifying genes.

Microsatellite primers are often developed in one species and used to assess neutral variability in related species. Such analyses may be confounded by ascertainment bias (i.e. a decline in amplification success and allelic variability with increasing genetic distance from the source of the microsatellites). In addition, other factors, such as the size of the microsatellite, whether it consists of perfect or interrupted tandem repeats, and whether it is autosomal or X-linked, can affect variation. To test the relative importance of these factors on microsatellite variation, we examine patterns of amplification and allelic diversity in 52 microsatellite loci amplified from five individuals in each of six populations of Cyrtodiopsis stalk-eyed flies that range from 2·2% to 11·2% mitochondrial DNA sequence divergence from the population used for microsatellite development. We find that amplification success and most measures of allelic diversity declined with genetic distance from the source population, in some cases an order of magnitude faster than in birds or mammals. The median and range of the repeat array length did not decline with genetic distance. In addition, for loci on the X chromosome, we find evidence of lower observed heterozygosity compared with loci on autosomes. The differences in variability between X-linked and autosomal loci are not adequately explained by differences in effective population sizes of the chromosomes. We suggest, instead, that periodic selection events associated with X-chromosome meiotic drive, which is present in many of these populations, reduces X-linked variation.

Hitchhiking phenomena and genetic recombination have important consequences for a variety of fields for which birds are model species, yet we know virtually nothing about naturally occurring rates of recombination or the extent of linkage disequilibrium in birds. We took advantage of a previously sequenced cosmid clone from Red-winged Blackbirds (Agelaius phoeniceus) bearing a highly polymorphic Mhc class II gene, Agph-DAB1, to measure the extent of linkage disequilibrium across ~40 kb of genomic DNA and to determine whether non-coding nucleotide diversity was elevated as a result of physical proximity to a target of balancing selection. Application of coalescent theory predicts that the hitchhiking effect is enhanced by the larger effective population size of blackbirds compared with humans, despite the presumably higher rates of recombination in birds. We surveyed sequence polymorphism at three Mhc-linked loci occurring 1·5–40 kb away from Agph-DAB1 and found that nucleotide diversity was indistinguishable from that found at three presumably unlinked, non-coding introns (β-actin intron 2, β-fibrinogen intron 7 and rhodopsin intron 2). Linkage disequilibrium as measured by Lewontin's D' was found only across a few hundred base pairs within any given locus, and was not detectable among any Mhc-linked loci. Estimated rates of the per site recombination rate ρ derived from three different analytical methods suggest that the amounts of recombination in blackbirds are up to two orders of magnitude higher than in humans, a discrepancy that cannot be explained entirely by the higher effective population size of blackbirds relative to humans. In addition, the ratio of the number of estimated recombination events per mutation frequently exceeds 1, as in Drosophila, again much higher than estimates in humans. Although the confidence limits of the blackbird estimates themselves span an order of magnitude, these data suggest that in blackbirds the hitchhiking effect for this region is negligible and may imply that the per site per individual recombination rate is high, resembling those of Drosophila more than those of humans.

Selective genotyping concerns the genotyping of a portion of individuals chosen on the basis of their phenotypic values. Often individuals are selected for genotyping from the high and low extremes of the phenotypic distribution. This procedure yields savings in cost and time by decreasing the total number of individuals genotyped. Previous work by Darvasi et al. (1993) has shown that the power to detect a QTL by genotyping 40–50% of a population is roughly equivalent to genotyping the entire sample. However, these power studies have not accounted for different strategies of analysing the data when phenotypes of individuals in the middle are excluded, nor have they investigated the genome-wide type I error rate under these different strategies or different selection percentages. Further, these simulation studies have not considered markers over the entire genome. In this paper, we present simulation studies of power for the maximum likelihood approach to QTL mapping by Lander & Botstein (1989) in the context of selective genotyping. We calculate the power of selectively genotyping the individuals from the middle of the phenotypic distribution when performing QTL mapping over the whole mouse genome.

Effective population size (Ne) is an important parameter determining the genetic structure of small populations. In natural populations, the number of adults (N) is usually known and Ne can be estimated on the basis of an assumed ratio Ne/N, usually found to be close to 0·5. In farm animal populations, apart from using pedigrees or genetic marker information, Ne can be estimated from the number N of breeding animals, and a value of 1 is commonly assumed for the ratio Ne/N. The purpose of this paper is to show the relation between effective population size and breeding herd size in livestock species. With overlapping generations, Ne can be predicted knowing the number of individuals entering the population per generation and the variance of family size, the latter being directly related to the survival pattern (or replacement policy) in the breeding herd. Assuming an ideal survivorship leading to a geometric age distribution, it can be shown that the number of breeding animals tends to overestimate effective size, particularly in early-maturing species. The ratio of annual effective size to the number of breeding animals is shown to be equal to [1+(a−1)(1−s)]2/(1−s2), where a is the age at first offspring and s is the survival rate (including culling) of the parents between successive births. This expression shows to what extent inbreeding may be determined by demography or culling policy independently of the actual herd size. In many situations a fast replacement or an early culling will increase annual effective size. Consequences for the management of small populations are discussed.

The statistical analysis of quantitative trait locus (QTL) experiments relies on the use of a linkage map of the markers genotyped. Such a map is, at best, a good estimate of the true map. Resources might be diverted into developing better marker maps or improved maps become available after the analysis, raising concerns over the original analysis. It is therefore important to understand the sensitivity of QTL analysis to map inaccuracy. We have used simulation methods to investigate the consequences of an incorrect map on the results of a QTL analysis using interval mapping. Backcross data sets were generated with a particular map and then analysed with both the correct map and incorrect maps. If the incorrect maps maintained the true linkage groups (i.e. no markers were incorrectly assigned to another linkage group), the accuracy of the map had little or no impact on the ability to detect QTLs, the true significance levels of the tests or the relative placement of QTLs. When a marker was incorrectly placed on another linkage group, there was a small increase in the level of the test. After adjusting for this increase, there was a decrease in power to detect a QTL near the misplaced marker. This decrease was of a similar magnitude to that found when using a single-marker analysis compared with interval mapping. These results mean that QTL analyses can proceed without the need for very accurate marker maps, and that estimated QTL positions can be translated onto updated maps without the need for reanalysis.

Genotype-assisted selection (GAS), i.e. selection for an identified quantitative trait locus (QTL) and polygenic background genes, has been shown to increase short-term genetic gain but may reduce long-term genetic gains. In order to avoid this reduction of long-term gain, multi-generation optimization of truncation selection schemes is needed. This paper presents a multi-generation optimization of optimum contribution (OC) selection with selection on an identified QTL. This genotype-assisted optimum contribution (GAOC) selection method assumes that the optimum selection differential at the QTL is constant over the time horizon, and achieves this by controlling the increase of the frequency of the positive QTL allele. Implementation was straightforward by an additional linear restriction in the OC algorithm. GAOC achieved 35·2%, 2·3% and 1·1%, respectively, more cumulative genetic gain than OC selection (ignoring the QTL) using time horizons of 5, 10 and 15 generations. When one-generation optimization of GAS was used instead of multi-generation optimization, these figures were 2·8%, 3·1% and 3·2%, respectively. Simulated annealing was used to optimize the increases of the frequency of the positive QTL allele in order to test the optimality of GAOC. This latter resulted in genetic gains that were always within 0·4% of those of GAOC. In practice, short-term genetic gains are also important, which makes one-generation optimization of genetic gain closer to optimal.

We investigated potential effects of parent-of-origin specific quantitative trait loci (QTL) in chicken. Two divergent egg-layer lines differing in egg quality were reciprocally crossed to produce 305 F2 hens. Searching the genome using models with uni-parental expression, we identified four genome-wide significant QTL with parent-of-origin effects and three highly suggestive QTL affecting age at first egg, egg weight, number of eggs, body weight, feed intake, and egg white quality. None of these QTL had been detected previously using Mendelian models. Two genome-wide significant and one highly suggestive QTL show exclusive paternal expression while the others show exclusive maternal expression. Each of the parent-of-origin specific QTL explained 3–5% of the total phenotypic variance, with the effects ranging from 0·18 to 0·4 phenotypic SD in the F2. Using simulations and further detailed analyses, it was shown that departure from fixation in the founder lines, grand-maternal effects (i.e. mitochondrial or W-linked) and Z-linked QTL were unlikely to give rise to any spurious parent-of-origin effects. The present results suggest that QTL with parent-of-origin specific expression are a plausible explanation for some reciprocal effects in poultry and deserve more attention. An intriguing hypothesis is whether these effects could be the result of genomic imprinting, which is often assumed to be unique to eutherian mammals.