These aquaculture- and conservation-oriented commentaries are not abstracts written by the original authors.  They reflect the opinions of someone else -- usually Roger Doyle.  Direct quotations from the papers or abstracts are marked with inverted commas.

137.  Outbred fish (population hybrids) are more resistant to ich
        Differences in initial and acquired resistance to Ichthyophthirius multifiliis between populations of rainbowfish. 2000. Gleeson, D.J., H.I. Mccallum, and I. P.F. Owens. Journal of Fish Biology 57 (2):466-475.
        Two pure species of rainbowfish (Melanotaenia) from three locations in Australia were experimentally infected with the parasite Ichthyophthirius multifiliis, a.k.a. "ich". One of the pure species, M. eachamensis, was much more susceptible to the parasite than the other species, M. splendida. Splendida served as a control for a follow-up hybridization experiment which involved crossing eachamensis from the original population with another population of the same species located some distance away. The population hybrids had significantly higher resistance than the single-population fish. "It is tentatively suggested that there may be a link between the heterozygosity of populations of rainbowfish and their initial ability to resist infection by Ichthyophthirius multifiliis." dgleeson@zoology.uq.edu.au

136.  Recombination distorts our understanding of the evolution of a viral epidemic
       Consequences of recombination on traditional phylogenetic analysis. 2000. Schierup, M.H., and J. Hein. Genetics 156: 879-891.
        When sequence data are analysed to generate a phylogenetic tree, the assumption is usually made that recombination within the sequence does not occur. Yet we know that this assumption is definitely wrong for viruses and for nuclear DNA in higher organisms, and possibly for mitochondrial DNA too.
        (Viruses can recombine when a host is simultaneously infected with more than one strain. It has been suggested that Asian integrated aquaculture of ducks, fish and pigs will soon bring an end to us all when avian and human influenza viruses recombine in the pig. The pig can catch both strains and pass the recombinant on to the farmer.)
        Recombination means that different parts of a sequence will have different evolutionary histories. The ground shifts: we are no longer trying to discover the one "true" tree for the sequence. The sequence is the outcome of a number of correlated trees.
        The authors of this paper used the same phylogenetic software most people use (PHYLIP) to investigate the effects of recombination. "With recombination present, the length of terminal branches and the total branch length are larger, and the time to the most recent common ancestor smaller, than for a tree reconstructed from sequences evolving with no recombination."
        In the context of this GCL website -- aquaculture and aquatic genetic conservation -- the paper may be most relevant to viral disease. The authors found that "The results for the [HIV and foot-and-mouth disease] viral data sets show strong evidence for recombination with very large estimated values. ... For HIV, [the analysis] also provides evidence for exponential growth, which agrees with the rapid spread of this virus. ... the rates appear here to be so high that phylogenetic analysis may be of very limited value."
        The authors conclude from their analysis of four data sets that "recombination may be sufficiently high to invalidate the use of phylogenetic trees in many population studies." Plainly, recombination needs to be taken into account if sequence data are used to draw epidemiological conclusions about the spread of aquacultural diseases, as suggested in October list #111. The computer program for calculating the various statistics is available at http://www.bioinf.au.dk/~mheide . The contact e-mail address is  mikkel.schierup@biology.au.dk 

135.  Virus attacks a protected species
        Canine virus blamed in Caspian seal deaths. 2000. Stone, R. Science 289 (5487): 2017-2018.
        The Caspian seal, listed as vulnerable by the World Conservation Union, may recently have lost as many as 20,000 individuals to an infection identified as canine distemper virus (CDV), i.e. a dog virus. The identification was made by a research group in Belfast working in collaboration with scientists from the former bio-weapons lab VECTOR, which is located upwind of Novosibirsk. VECTOR believes that the situation is even more serious, and that the deaths are due to simultaneous infection by CDV and an influenza virus in the seals [where the viruses recombine? -- just kidding]. For more on this development see the September-October publication from the Belfast group led by Seamus Kennedy at  http://www.cdc.gov/ncidod/eid/vol6no6/kennedy.htm  E-mail  seamus.kennedy@dardni.gov.uk

134.  Hybridization of locally-adapted finches promotes adaptive radiation
        The mitochondrial and nuclear genetic homogeneity of the phenotypically diverse Darwin's groundfinches. 1999. Freeland, J.R., and Boag. P.T. Evolution 53 (5):1553-1563.
        The drab-looking finches of the Galapagos islands, which turned Charles Darwin towards the discovery of natural selection, are among the best-studied and best-understood examples of this evolutionary process. In this paper the authors describe the first successful attempt to resolve the evolutionary history and genetic relationships of the birds. It is not surprising that the paper is relevant to issues in genetic conservation and even aquaculture.
        The authors found that "The differentiation of the ground finch species based on morphological data is not reflected in either mitochondrial or nuclear DNA sequence phylogenies." Not only that, inferred genealogies based on mitochondrial and nuclear markers are not even concordant with each other. The authors suggest that "the absence of species-specific [DNA sequence] lineages can be attributed to ongoing hybridization involving all six species of Geospiza." Hybrids have an advantage in some years so there is no selection against hybridization even though there is strong, ongoing selection for morphology (e.g. the size of the beak in relation to the size of the seeds available for food). "Hybridization has apparently played a role in the adaptive radiation of Darwin’s finches."
        There is a lot of evidence for local adaptation of migratory salmonid populations (e.g. October #115), but those who argue that this will result ipso facto in outbreeding depression might ponder Darwin's finches. We need to have direct evidence for outbreeding depression when salmonids hybridize between spawning areas. Then we can know for sure whether hybridization is an evolutionary problem, rather than an evolutionary solution, for salmonids at risk of extinction.  j.r.freeland@reading.ac.uk

133.  The genetics (?) of acquired (?) tolerance to shrimp WSSV
        Quasi-immune response of Penaeus japonicus to penaeid rod-shaped DNA virus. 2000. Venegas, C.A. , L. Nonaka, K. Mushiake, T. Nishizawa, and K. Murog. Diseases of Aquatic Organisms 42 (2):83-89.
        The authors experimentally infected Kuruma prawns with white spot syndrome virus (WSSV). The shrimp that survived this experimental challenge, as well as the survivors of a natural outbreak of WSSV, demonstrated high levels of survival to subsequent experimental challenges by injection of the virus. PCR testing showed that many of the survivors continued to carry the virus even though they were apparently healthy. The paper refers to recent work in Thailand by Flegel and others which also may show induction of tolerance to WSSV and yellowhead virus YHV within a single generation of shrimp following massive mortalities.
        The authors of this carefully-designed experiment conclude that the "high survival rates suggested that survivors (natural or experimental) were able to resist infection and that the resistance was not due to selection of naturally resistant [or tolerant?] shrimp ... but due to enhancement of an immune-like system (quasi-immune response) after exposure to [the virus]."
        Whether to tag this phenomenon as "tolerance" or "resistance" is not a mere quibble. The two processes have very different implication for the long-term evolution of a host/pathogen system. See for example September #105, August #96, July #84. My highly non-expert interpretation of this experiment is that surviving animals carry the virus but aren't sick, therefore both components survive indefinitely in the aquaculture system, which means tolerance in the host. From the long-term genetic perspective the evolutionary implications of this may be good news for the industry.
        Also from the genetics perspective, one is tempted to speculate about the variation in the proposed quasi-immune response. Is the partial survival at the first (inducing) challenge due to variation in the strength of the response? Or in the threshold for induction or enhancement? Or is there any selectable genetic variation at all in the shrimp (the details of the experimental design suggest that the environment is involved in a complicated way)? The answers to such questions will profoundly influence the design of genetic improvement programmes. fpath@hiroshima-u.ac.jp

132.  Inbreeding depresses fitness in the wild
        Inbreeding depression influences lifetime breeding success in a wild population of red deer (Cervus elaphus). 2000. Slate, J., L.E.B. Kruuk, T.C. Marshall, J.M. Pemberton, and T.H. Clutton-Brock. Proceedings of the Royal Society (B) 267 (1453):1657-1662.
        The authors used microsatellite heterozygosity as an indicator of individual inbreeding coefficients among unmanaged deer on the island of Rum, Scotland. They found that heterozygosity is correlated with lifetime breeding success (total offspring) in both males and females. They believe that this is the first observation of such a correlation between inbreeding and something close to overall fitness, in both sexes. Similar correlations have previously been found between heterozygosity/inbreeding and major components of fitness such as mating success (July #82). Lifetime breeding success is not quite identical with fitness measures such as r or R₀ which might be calculated from a life table or a Leslie matrix. Both total production and the schedule of the production of offspring (as a function of overall population growth rate) are important to fitness. j.pemberton@ed.ac.uk

131.  Inbreeding depression in the wild no worse than in the lab
        Equivalent inbreeding depression under laboratory and field conditions in a tree-hole-breeding mosquito. 2000. Armbruster, P., R.A. Hutchinson, and T. Linvell. Proceedings of the Royal Society (Biology) 267 (1456):1939-1945.
        The authors expected to find that inbreeding had a more profound depressive effect on fitness in the natural environment than in an optimized laboratory environment where animals are well taken care of. This is a reasonable expectation based on theory as well as many recent observations. However, in this case, they found that while the fitness of the mosquito populations did decline with inbreeding, the decline was the same in natural tree-holes as under favourable laboratory conditions. A 10% increase in inbreeding coefficient reduced fitness, as measured by R₀ an index which is essentially the absolute rate of population growth calculated from the life table (see #132, above) , by 12%-15%. Thus we can not be totally confident in predicting that inbred small and endangered populations (e.g. of fish, by analogy) must inevitably be in worse shape than predicted by inbreeding experiments in the laboratory, aquarium or farm.
        The authors suggest that a composite fitness index may be a more appropriate measure of inbreeding depression than individual component of fitness such as fecundity or survival. If trade-offs between fitness components occur this may well be true. Note September list #101 in which inbreeding depression was found to be specific to each trait, not a generalized phenomenon. parmbrus@zoo.uvm.edu

130.  Escaped farmed salmon are unfit in Norway -- and their offspring too
        Lifetime success and interactions of farm salmon invading a native population. 2000. Fleming, I.A., K. Hindar, I.B. Mjølnerød, B. Jonsson, T. Balstad, and A. Lamberg. Proceedings Royal Society (UK), Ser. B 267 (1452):1517-1523.
        The authors report the result of an experiment in which mature, farmed salmon were released into the Imsa river in Norway. Radio tagging and genetic analysis revealed that "The farm fishes were competitively and reproductively inferior, achieving less than one-third the breeding success of the native fishes." Most of the gene flow involved native males mating with farmed females -- farmed males were relatively uncompetitive. "There were also indications of selection against farm genotypes during early survival but not thereafter. ... Ultimately, the lifetime reproductive success (adult to adult) of the farm fishes was 16% that of the native salmon." The authors note, however, that the invasion of the Imsa by farmed fish reduced the productivity of the native population by more than 30% through resource competition and competitive displacement. ian.fleming@ninatrd.ninaniku.no

129.  The right way to estimate quantitative genetic parameters in nature
        Estimating variance components and heritabilities in the wild: a case study using the ‘animal model’ approach. 2000. Milner, J.M., J.M. Pemberton, S. Brotherstone, and S.D. Albon. Journal of Evolutionary Biology 13 (5):804-813.
        The authors of this paper have published a fine example of a type of study that a lot of people are attempting -- estimation of quantitative genetic parameters in wild populations. In this case the population was the Soay sheep living on the island of St. Kilda, and the techniques included markers to estimate first-order genetic relationships (the relationship matrix) which were then used for the modern animal model technique for parameter estimation. The trait that experienced the strongest selection was body weight. This trait also had both the highest additive variation and the lowest heritability. jos.milner@ic.ac.uk

128.  Prawn markers and biogeography in the Pacific
        Population structure of the giant tiger prawn Penaeus monodon in Australian waters, determined using microsatellite markers. 2000. Brooker, A.L., J.A.H. Benzie, D. Blair, and J.-J. Versini. Marine Biology 136 (1):149-157.
        The authors used microsatellite polymorphisms to discover that the monodon populations in western Australia are genetically quite different (and less variable) than those on the north and east. "Reduced variability is consistent with a recent population bottleneck or colonization by a small founding population from the east when sea links between Indonesia, New Guinea and Australia were re-established following the last ice age." Good microsatellite markers have been rather hard to find in shrimp, for technical reasons, and these look like winners: 34, 35 and 85 alleles! Amanda.Brooker@jcu.edu.au

127.  BLUP selection increases inbreeding
        Prediction of rates of inbreeding in populations selected on best linear unbiased prediction of breeding value. 2000. Bijma, P., and J.A. Woolliams. Genetics 156:361-373.
        The best linear unbiased predictor (BLUP) combines information from relatives to maximise the rate of genetic gain from selection. BLUP is now being used for breed improvement in aquaculture and, with the advent of marker-based reconstruction of pedigrees, related techniques are used to estimate quantitative genetic parameters in natural populations. "Predictions for the rate of inbreeding (F) in populations with discrete generations undergoing selection ... were developed. ... Expected contributions of individuals were predicted using a linear model, ui(x) = + ßsi, where si denotes the selective advantage as a deviation from the contemporaries, which was the sum of the breeding values of the individual and the breeding values of its mates." .... [Simulation showed] that, contrary to random selection, F less than halved when the number of parents was doubled and that in specific cases F may increase with the number of dams."
        The authors provide a useful rule of thumb for dealing with this situation, which may have considerable importance in aquaculture and, possibly, in genetic conservation of fish, where artificial selection might be performed to reduce domestication effects in the hatchery. "The cost of raising offspring from parents that are destined not to contribute [genes] in the long term is a hidden cost of the high rates of inbreeding associated with BLUP selection." piter.bijma@alg.vf.wau.nl

126.  PCR shrimp for virus without killing them
         A non-destructive method based on the polymerase chain reaction for detection of hepatopancreatic parvovirus (HPV) of penaeid shrimp. 2000. Pantoja, C.R., and D.V. Lightner. Diseases of Aquatic Organisms 39:177-182.
        "Current methods to detect hepatopancreatic parvovirus (HPV) infection of penaeid shrimp depend on invasive techniques that require dissecting the organs infected by this virus. ... A method was developed for HPV detection by applying a polymerase chain reaction (PCR) assay to fecal samples collected from live HPV-infected shrimp Penaeus chinensis. ...Analysis of fecal samples by PCR may prove useful for non-lethal screening of valuable shrimp [e.g. broodstock shrimp] of unknown HPV status."
        Presumably the techniques could also be useful for distinguishing and then selecting resistant as opposed to tolerant animals. cpantoja@u.arizona.edu

125.  How to measure growth in genetic experiments
        Statistical models for estimating the genetic basis of repeated measures and other function-valued traits. 2000. Jaffrézic, F., and S.D. Pletcher. Genetics 156:913-922.
        Growth rate, one of the prime targets for genetic improvement in aquaculture as well as for QTL mapping, is not easy to measure in a meaningful way. In fact the definition of growth gets more complex -- mushy -- the more you worry about it. Similar problems arise with other important traits that are measured repeatedly as a function of another variable: fecundity vs. age, mating success vs. size, size vs. temperature, survival vs. salinity, etc.
        There is a choice of three mathematical models that are sometimes used for this purpose: regression, approximation of the data by orthogonal polynomials, and character process models. The discussion of the contrasting features of the 3 approaches is admirably clear in this paper. The authors elaborate on the third approach (which they find to be the best) and evaluate all of them using both simulated and real data. All models were estimated from data by maximum likelihood (REML). The data consisted of a genetic design (sire half-sib families) that is standard in experimental quantitative genetics and achievable in some natural populations using genealogies reconstructed from DNA markers.
        "The simplicity of the character process model allows quantitative statements about the predominant attributes of the genetic covariance function. Genetic variance for Drosophila mortality declines significantly with age, while genetic variance is constant at all ages for reproductive output." The authors also analysed growth in beef cattle. They are making the software available for free. s.pletcher@ucl.ac.uk