The Bayesian controversy in animal breeding

Frequentist and Bayesian approaches to scientific inference in animal breeding are discussed. Routine methods in animal breeding (selection index, BLUP, ML, REML) are presented under the hypotheses of both schools of inference, and their properties are examined in both cases. The Bayesian approach is discussed in cases in which prior information is available, prior information is available under certain hypotheses, prior information is vague, and there is no prior information. Bayesian prediction of genetic values and genetic parameters are presented. Finally, the frequentist and Bayesian approaches are compared from a theoretical and a practical point of view. Some problems for which Bayesian methods can be particularly useful are discussed. Both Bayesian and frequentist schools of inference are established, and now neither of them has operational difficulties, with the exception of some complex cases. There is software available to analyze a large variety of problems from either point of view. The choice of one school or the other should be related to whether there are solutions in one school that the other does not offer, to how easily the problems are solved, and to how comfortable scientists feel with the way they convey their results.     

Modifying growth curve parameters by multitrait genomic selection

The use of genomic selection for changing the growth curve shape of animals, acting simultaneously on the 3 parameters of a Gompertz growth curve, was studied using computer simulation. Results showed that genomic selection modified the growth curve. Responses to all selection criteria were accompanied with a correlated response in mature weight due to the high genetic correlations between mature weight and the 2 rate parameters. Responses to selection were affected by the loss of accuracy over generations of genomic selection due to the loss of gametic disequilibrium between SNP and genes determining the parameters. In conclusion, genomic selection can be used for changing growth curve parameters. However, changing the whole curve (birth weight, adult BW, and curve shape) is not easy due to the correlation between all BW along the growth of the animal. Applying genomic selection will require a constant reevaluation of the associations between SNP and genes determining the curve parameters.     

STUDY OF THE INTERACTIONS BETWEEN IODINE AND MINERAL NUTRIENTS IN LETTUCE PLANTS

Iodine is vital to human health, and iodine bio-fortification programs help improve the human intake through plant consumption. Essential processes in plants that can be altered by iodine bio-fortification include the nutritional state of plants. The main objective of this work was to determine whether mineral nutrient concentrations were affected by the application of different forms (iodide [I^?] vs. iodate [IO₃ ^?]) and dosages (0, 20, 40, and 80 μM) of iodine, to ascertain the influence of this trace element in Lactuca sativa var. longifolia plants. The application of 80 μM of I^? significantly reduced the nutrients nitrogen (N), phosphorus (P), and potassium (K) to below the optimal ranges established for this crop. The IO₃ ^? treatments represented optimal maintenance of the nutritional state for most of the nutrients and even an improvement of nutrients as important as magnesium (Mg) and iron (Fe).     

The effect of divergent selection for uterine capacity on fetal and placental development at term in rabbits: maternal and embryonic genetic effects

The aim of this work was to study the effects of the genotype of the dam, the embryo, or their interactions on prenatal growth by performing double-reciprocal embryo transfers between two lines of rabbits divergently selected for uterine capacity. Females from high (n = 53) and low (n = 48) lines were slaughtered at 72 h of gestation, and recovered embryos were transferred to the oviducts of recipient does from the high (n = 23) and low (n = 19) lines. Each recipient doe received eight embryos from the high line into one oviduct and eight embryos from the low line into the other. Recipient does were slaughtered on d 28 of gestation. The percentages of live fetuses at 28 d of gestation were 89.2 and 74% for high and low recipient lines, respectively. Length and weight of the empty uterine horn and weight of the full uterine horn were not affected by either the recipient or by donor line. Fetal weight was affected by the recipient line but not by the donor line. Fetuses gestated in high recipient does were 7% heavier (P < 0.10) than those in the low recipient does. There was a donor and a donor x recipient interaction effect on fetal placental weight. Fetal placental weight was heavier (7%, P < 0.01) for embryos from the low line. Embryos from the high line gestated in low-line uteri showed a lower fetal placenta weight than did low-line embryos gestated in high-line uteri and low-line uteri (P < 0.05). Linear regression coefficients of fetal weight at term on fetal placental weights differed (P < 0.05) for the high and low donors (4.33 +/- 0.28 and 3.41 +/- 0.29 respectively). A significant effect of the donor genotype on individual placental length was observed (P < 0.05), which might have resulted from a smaller individual placental length of low-line embryos gestated high-line uteri (P < 0.10). Neither donor nor recipient lines affected maternal placental weight or available space for fetuses. Fetuses and their fetal placentae were heavier when receiving more than four blood vessels than when receiving less than three blood vessels (13 and 17% respectively, P < 0.05). Neither recipient nor donor genotype affected the number of blood vessels arriving at each live fetus. Thus, fetal weight depends on the maternal genotype, whereas fetal placental weight depends on the embryo genotype in these two lines of rabbits divergently selected for uterine capacity.     

Selection for ovulation rate in rabbits: genetic parameters and correlated responses on survival rates

The aim of this work was to evaluate the correlated responses on survival rates after 10 generations of selection for ovulation rate (OR). Selection was based on the phenotypic value of ovulation rate estimated at d 12 of second gestation by laparoscopy. Traits recorded were litter size (LS), estimated as total number of rabbits born per litter in up to 5 parities; OR, estimated as the number of corpora lutea in both ovaries; the number of implanted embryos (IE), estimated as the number of implantation sites; the number of right and left IE (RIE and LIE); ovulatory difference (OD), defined as the difference between the right and the left OR, expressed as an absolute value; implantatory difference (ID), defined as the difference between RIE and LIE, expressed as an absolute value; embryonic survival (ES), calculated as IE/OR; fetal survival (FS), calculated as LS/IE; prenatal survival (PS), calculated as LS/OR. A total of 1,081 records were used to analyze ES, and 770 were used to analyze FS and PS. The number of records used to analyze the other traits ranged from 1,079 for ID to 3,031 for LS. Data were analyzed using Bayesian methodology. Genetic parameters of OR, OD, and LS were estimated in a previous paper. Estimated heritabilities of IE, ID, ES, FS, and PS were 0.11, 0.03, 0.09, 0.24, and 0.14, respectively. Estimated repeatabilities of IE, ID, and ES were 0.22, 0.12, and 0.20. Estimated phenotypic correlations of OR with ES, FS, and PS were -0.07, -0.26, and -0.28, respectively. Their estimated genetic correlations with FS and PS were negative (probability of being negative 1.00 and 0.98, respectively). Nothing can be said about the sign of the genetic correlation between OR and ES. Ovulation rate was phenotypically uncorrelated with ID. Their estimated genetic correlation was positive (probability of being positive 0.91). The genetic correlation of ID with PS and LS was not accurately estimated. Phenotypic and genetic correlations between LS and survival rates were positive (probability of being positive 1.00). In 10 generations of selection, FS decreased around 1% per generation. No correlated response in ES was observed. In summary, the decrease in FS in rabbits selected for OR seemed to be responsible for the lack of correlated response observed in LS.     

Effects of supercritical carbon dioxide (SC-CO(2)) oil extraction on the cell wall composition of almond fruits.

Extraction of oil from almond fruits using supercritical carbon dioxide (SC-CO(2)) was carried out at 50 degrees C and 330 bar on three sets of almonds: raw almond seeds, raw almond kernels, and toasted almond seeds. Three different oil extraction percentages were applied on each set ranging from approximately 15 to 16%, from approximately 27 to 33%, and from approximately 49 to 64%. Although no major changes were detected in the fatty acid composition between fresh and partially defatted samples, carbohydrate analysis of partially defatted materials revealed important changes in cell wall polysaccharides from almond tissues. Thus, at low extraction percentages (up to approximately 33%), pectic polysaccharides and hemicellulosic xyloglucans were the main type of polymers affected, suggesting the modification of the cell wall matrix, although without breakage of the walls. Then, as supercritical fluid extraction (SCFE) continues and higher extraction rates are achieved (up to approximately 64%), a major disruption of the cell wall occurred as indicated by the losses of all major types of cell wall polysaccharides, including cellulose. These results suggest that, under the conditions used for oil extraction using SC-CO(2), fatty acid chains are able to exit the cells through nonbroken walls; the modification of the pectin-hemicellulose network might have increased the porosity of the wall. However, as high pressure is being applied, there is a progressive breakage of the cell walls allowing the free transfer of the fatty acid chains from inside the cells. These findings might contribute to providing the basis for the optimization of SCFE procedures based on plant food sources.     

Variations on stress defences and metallothionein levels in the Senegal sole, <Emphasis Type="Italic">Solea senegalensis</Emphasis>, during early larval stages

The biochemical status of antioxidant defences of larvae from the commercial, benthic fish, Solea senegalensis, were studied over a period of 28days from hatching. The parameters studied were: catalase (EC 1.11.1.6), total glutation peroxidase activity (sum of the selenium dependent and independent forms) (EC 1.6.4.2), DT-diaphorase (EC 1.6.99.2), and as an indication of conjugative detoxifying metabolism, the enzyme glutathione S-transferase (GST; EC 2.5.1.18). Oxidative damage was evaluated by the formation of malondialdehyde (MDA) as an indication of lipid peroxidation (LP). Metallothionein (MT) levels were measured by differential pulse polarography and stress proteins (HSP 70 and HSP 60) were detected by immunoblotting, using commercial antibodies. The presence of catalytic activities was observed from hatching day and tended to increase during larval development. Significant changes were seen in most enzymes at day 3, when the larvae finished their endogenous feeding and started to feed on rotifers. DT-diaphorase activities were similar in both NADH and NADPH-dependent forms and, in turn, were about 10-times lower than their correspondent reductases. LP sharply increased at 19 and 28days post-hatch (dph), suggesting a saturation of the antioxidant defences during metamorphosis. MT showed high values during the endogenous phase and the lowest value corresponded to day 3, when the egg-yolksac was fully reabsorbed; a steady-state value was reached thereafter. Stress proteins were present at all stages, showing distinctiveness in the molecular weight of HSP60 at hatching day and 1dph. Quantitatively, both forms were significantly elevated at 3dph, with their ratio (HSP70/HSP60) decreasing over time.     

Contribution to the Development of the Stapedius Muscle Structure in Human Fetuses

The purpose of this report is to contribute to the knowledge of the morphogenesis of the human stapedius muscle. We have studied the stapedius structure development in 17 temporal bones from human fetuses. Chronological ages were from 48 days (21 mm) to 29 weeks (270 mm). Samples were fixed in 10% formol, decalcified with 2% nitric acid, embedded in Paraplast, sectioned in sequence of 7 microns and were stained with haematoxylin & eosin and Martins' trichrome. Four maturation stages were observed in the development of the stapedius muscle structure. First, the formation of the 'primordium muscularis' in the mesenchyme of the tympanic cavity (promyoblasts) was observed in a 36-mm embryo. Second, its differentiation into myoblasts, and its next cellular fusion to form muscle tubes, was seen in a 99-mm foetus. In the third stage, myofibrogenesis phenomena were developed, to synthesize, the myotube myoblasts, myofilaments, phenomena we observed that took place during the 14th to 17th weeks of development. The last stage was characterized by the presence of striated muscle fibres, at first intermingled with myotubes (18th to 26th weeks), until in the 29th week fetuses (270 mm) the skeletal muscle fibre was completely differentiated. We conclude that in the 29th week of development (270 mm) the stapedius muscle structure is completely differentiated, although the functional development is not complete until birth.