| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Institut National de la Recherche Agronomique, France,
and
* Station de Génétique Quantitative et Appliquée, 78352 Jouy-en-Josas Cedex;
and
Laboratoire de Génétique Cellulaire, 31326 Castanet Tolosan Cedex;
and
LABOGENA, 78352 Jouy-en-Josas Cedex;
and
Station Expérimentale de Sélection Porcine, 86480 Rouillé;
and
¶ Laboratoire de radiobiologie et d’étude du génome, 78352 Jouy-en-Josas Cedex; and
and
# Unité Mixte de Recherches sur le Veau et le Porc, 35590 Saint Gilles
3 Correspondence:
phone: +33 1 34 65 22 84; fax: +33 1 34 65 22 10; E-mail:
bidanel{at}dga.jouy.inra.fr.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
Key Words: Androstenone • Boar Taint • Gene Mapping • Pigs • Quantitative Trait Loci
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
An experiment has been conducted at the French National Institute of Agronomic Research to map loci affecting economically important traits in a three-generation experimental cross between Meishan (MS) and Large White (LW) pig breeds. The large differences reported between these two breeds for fat androstenone levels, along with differences for sexual maturity (Prunier et al., 1987), makes it likely to have a number of genes segregating in a Meishan x Large White F2 population contributing to the variation of these traits.
The objective of this research was to map QTL affecting fat androstenone concentrations in pigs. Because of the large breed differences in sexual maturity, fat androstenone levels were measured at several stages of life in F2 intact males. The usual line-cross (LC) analyses were complemented with systematic tests for imprinting and linked QTL. The possible segregation of QTL alleles in founder populations was also investigated.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
The F1 sows were managed under a batch farrowing system, with a 3-wk interval between contiguous batches. These batches then became postweaning and fattening batches of growing pigs. Piglets were weaned at 4 wk of age and placed in a postweaning unit until 10 wk of age. Male piglets were not castrated and were transferred at 10 wk of age to another INRA experimental herd at Rouillé, Vienne.
When arriving in Rouillé, male piglets were allotted to pens of about 10 animals in a semiopen building. They were given ad libitum access to a diet containing 17% crude protein, 0.85% lysine, and 3,100 kcal digestible energy during the whole testing period from 10 to 22 wk of age. They were slaughtered when they reached 90 kg live weight or 180 d of age at a French National Institute of Agronomic Research experimental slaughterhouse located in Saint-Gilles, Ille-et-Vilaine. Mean and standard deviation of weight at slaughter were, respectively, 79 and 13 kg.
A sample of intact male pigs from this experimental F2 population, described in Bidanel et al. (2001a), was utilized for the present study. It consisted in 485 F2 males derived from 24 full-sib families (Table 1
).
|
The number of records, overall means, and standard deviations of the five traits studied are shown in Table 2. Phenotypic correlations between traits are shown in Table 3. Number of records varied according to the trait due to the accidental loss of some of the backfat samples. The lower number of samples at 160 d of age results from a number of animals being slaughtered before they reached that age.
|
|
Statistical Methods
The analyses were performed on the natural logarithm of fat androstenone measurements in order to normalize their distribution. Table 2
also shows overall means and standard deviations of log-transformed data. Sex-average distances were used in all analyses for comparison purposes, as Knott et al. (1998) showed that using sex-specific maps had limited effects on the results.
Two types of interval mapping analyses were performed: 1) a LC analysis assuming that founder populations are fixed for alternative QTL alleles and 2) a model assuming that the F2 population is a mixture of full- and half-sib (HFS) families and making no assumptions about the number of QTL alleles and their frequencies within the founder populations (referred to as HFS model hereafter).
Line-Cross Model.
The LC model used assumed a diallelic QTL with alternative alleles fixed in founder breeds, Q in Meishan and q in Large White animals. Denoting the effects of QQ, Qq/qQ, and qq as a, d, and -a, respectively, the performance of an F2 offspring ij in the ith contemporary group can be written as:
 | 1 |
where yij is the phenotype of the ijth individual, gi is the fixed effect for the ith fattening batch, covij is the covariate age for A100, A120, A140, and A160 and weight at sacrifice for A_SL, caij, and cdij are the coefficients of additive and dominance components, respectively, for animal ij at any putative location in the genome, and eij is the residual error. Coefficients caij and cdij were computed as caij = probij(QQ) - probij(qq) and cdij = probij(Qq) + probij(qQ), where probij(XX) is the probability of animal ij having the genotype XX. The genotypic probabilities were computed as described in Haley et al. (1994) considering only the most probable phases. At each location (each centimorgan), an F-ratio was computed comparing model {1}with one QTL to an equivalent model without any linked QTL. Estimates for a and d were calculated at the location with the highest F-ratio.
Additional LC analyses were carried out to test the presence of linked QTL, of parent-of-origin effects and of family x QTL interactions, as described by Knott et al. (1998) and Quintanilla et al. (2002). The presence of two QTL in the same linkage group was tested by adding additive and dominance effects for a second QTL in the model and carrying out a two-dimensional search. Parent-of-origin effects were tested by considering the paternal or maternal origin of grandparental (MS or LW) alleles, including the difference between the two classes of heterozygotes in the model.
Most analyses were done using the software developed by Seaton et al. (2002). The different hypotheses were tested by computing, at every centimorgan of the whole genome, the reduction in sum of squares (F-ratio test) caused by adding the new component(s) to a no-QTL and to a single QTL models. Parameter estimates were calculated at the location with the highest F-ratio.
Significance thresholds at chromosome- and genome-wide levels were determined empirically as described in Quintanilla et al. (2002), by data permutation (Churchill and Doerge, 1994) and Bonferroni correction (Knott et al., 1998). Values considered as genomewide suggestive thresholds (1 false positive per genome scan) were, respectively, F = 8.0, 5.9, 4.7, 4.1 for F-ratios with one, two, three, and four degrees of freedom in the numerator. The corresponding genomewide significant thresholds (0.05 false-positives per genome scan) were, respectively, F = 13.5, 9.0, 6.1, and 5.5.
Half- and Full-Sibs Model.
Only a single QTL model was considered in the HFS approach. Logarithms of fat androstenone levels were adjusted for fattening batch and age or weight at measurement, using a fixed linear model. Test statistics were computed, at any putative location on the chromosome, as the ratio of likelihoods under the hypotheses of one (H1) vs no QTL linked to the set of markers considered. Under H1 hypothesis, a QTL with a gene substitution effect (
) for each sire and each dam (within sire) was fitted to the data (Le Roy et al., 1998; Bidanel et al., 2001a). After maximization of this function, average substitution effects (), which in the present case are equivalent to additive values (a), were hence estimated within each sire (half-sib) family and within each dam (full-sib) family as described in Bidanel et al. (2001a) and averaged over families.
Significance thresholds were determined empirically by simulating the data assuming a polygenic infinitesimal model and a normal distribution of performance traits as described by Le Roy et al. (1998) and Bidanel et al. (2001a). Estimated thresholds ranged from 53.8 to 56.9 and from 65.1 to 70.3, respectively, for suggestive and significant genomewide linkages.
Candidate Gene Analyses
Based on QTL results, two loci coding for enzymes involved in 16-androstenes and/or androstenone pathways were investigated as positional candidate genes: the steroid chromosome P450 21-hydroxylase (CYP21) and cytochrome P450 cholesterol side chain cleavage subfamily XIA (CYP11A) loci.
Sequencing of CYP21 Gene.
Primers amplifying the sequence of CYP21 were designed using Primer3 (Rozen and Skaletsky, 1998) from available porcine sequence (GenBank Accession no M83939) and are described in Table 4. All fragments were amplified in the same conditions in a volume of 25 µl containing 50 ng of DNA, 200 mM dNTP, 0.5 µm primers, 1.5 mM MgCl2, 1x buffer and 0.5 unit of Taq polymerase (Gibco BRL, Invitrogen Corporation, Carlsbad, CA) on a GeneAmp System 9700 (Applied Biosystems, Foster City, CA). Amplification of fragments was performed by denaturation of DNA at 95°C during 5 min, followed by 30 cycles of amplification consisting in 30 s at 95°C, 30 s at 58°C, 45 s at 72°C, and a final elongation of 15 min at 72°C. The PCR fragments were purified with QIAquick PCR Purification Kit (QIAGEN Inc., Valencia, CA) and sequenced on the ABI 3700 by using BigDye Terminator RR mix (Applied Biosystems). The reaction sequence was performed after 5 min of DNA denaturation at 95°C during 25 cycles (30 s at 95°C, 15 s at 58°C, and 4 min at 60°C) with a mix of 15 µL of purified PCR product, 4 µL of BigDye, and 3.5 pmol of primer. The sequences of MS and LW alleles were submitted to Genbank (AF490410).
|
). The CYP11A locus was mapped using the 118 clones of IMpRH panel (Yerle et al., 1998), and PCR fragments were analyzed on 2% agarose gels. The vector of result was analyzed using IMpRH Server (Milan et al., 2000). Primers designed by Greger (2000) were used to amplify a DNA fragment containing the candidate mutation CYP C/T -155. For an easy screening of this mutation, a PCR-RFLP test with an internal digest control was developed using NLAIII enzyme. After amplification of the fragment using conditions described for CYP21 gene (but with an annealing temperature of 60°C), 10 µL of PCR fragments were subjected to digestion for 2 h at 37°C in a volume of 20 µL containing 5 units of NLAIII enzyme (Gibco BRL) and 1x buffer supplied by the manufacturer. Digestion of the T-allele produced two fragments (446 bp and a control fragment of 189 bp), whereas digestion of the C-allele produced three fragments (160 bp and 286 bp and the control fragment of 189 bp). Fragments were analyzed on 2% agarose gels.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
. Three chromosomal regions located on Sus Scrofa chromosomes (SSC) 3, 7, and 14 reached genomewide significance for at least one measure of androstenone level using both LC and HFS models. Significant QTL were also detected on SSC 4 and 9 with the LC model, but only suggestive (SSC 4) or nonsignificant (SSC 9) results were obtained using HFS model. Plots of F-ratios across SSC 3, 4, 7, and 14 are shown in Figure 1.
|
|
Strong evidence of a QTL was also found on SSC 3, though LC and HFS models gave somewhat different significance levels. The F-statistics (LC model) reached a significant level for three androstenone measurements (A100, A120, and A140) and a suggestive level for A_SL, whereas the likelihood ratio (HFS model) was highly significant only for A140 and suggestive for A120. Meishan alleles all had positive additive effects explaining from 7 to 11% of the phenotypic variance.
On SSC 4, 9, and 14, significant QTL were detected with the LC model for a single fat androstenone measurement, and suggestive QTL were obtained for all, or part of, the other ones. Meishan alleles were in all cases associated with high androstenone levels and had completely additive effects, except for SSC4, where LW alleles showed complete dominance over MS alleles for A100, A120, and A140. The HFS model gave results similar to LC model on SSC 14, with identical location of the QTL (in the S0058 to S0007 interval) but somewhat lower estimated additive effects. Less consistent results were obtained for SSC 4: HFS model gave suggestive results for only three traits, and different most likely positions (104 vs 64 cM) were obtained for the QTL affecting A160. Conversely, no evidence of a QTL was found on SSC 9 with the HFS model.
Suggestive QTL were detected on SSC 13 but with rather inconsistent results between LC and HFS models. Traits affected, QTL positions and estimated effects varied according to model used. A QTL affecting A120, A160, and A_SL was detected in the SWR1941 to S0222 interval with the LC model. The QTL explained 4 to 9% of trait variance, with a positive dominant effect of MS alleles. Conversely, a unique QTL affecting A140 was suggested at the end of the q arm of SSC 13 using HFS model, with a close to zero average allele substitution effect.
Five additional chromosomal regions located on SSC 2, 6, 10, 11, and 18 also reached a suggestive level of significance for one trait x model combination.
Additional Analyses
No evidence of linked QTL was found. Nevertheless, the profile of F-ratio for SSC 7 and 4 in the single QTL analysis (see Figure 1) seemed to indicate two almost equally probable locations. Moreover, the two-QTL analysis on SSC 7 (data not shown) gave highly significant results for the test of two QTL vs no QTL, and the improvement of fit obtained by adding a second QTL vs the best single QTL approached the suggestive level of significance (the F-ratio with 2 df in the numerator ranged from 4.58 to 5.20). Conversely, results obtained in the two-QTL analysis on SSC 4 were far from reaching the suggestive level.
No evidence of imprinted QTL affecting fat androstenone levels could be found across the Sus scrofa genome. Similarly, no QTL by family interaction effect approached significance at any position on the whole genome.
Fat androstenone levels were positively correlated with male sexual development traits measured at slaughter (see Table 6), particularly with bulbo-urethral gland and seminal vesicles that are target organs for androgens and estrogens.
|
Radiation hybrid mapping of CYP11A led to a position close to marker SWR1210 (LOD score of 18.9), between S0102 and SW352. The 12 founder pigs and the F1 boars were then analyzed for the CYP C/T-155. The frequency of the T-allele was 33% (4/12) and 8% (1/12) in MS and LW founder pigs, respectively. Only one F1 male was found to be heterozygous for the C/T mutation, the five other boars being homozygous C/C. Moreover, the C/T heterozygous male inherited its T-allele from the LW line.
The CYP C/T-155 mutation was also analyzed on a set of 62 Large White x (Landrace x Large White) crossbred pigs, for which blood samples and androstenone measurements were available (Sellier et al., 1987). Results are shown in Table 7. The T-allele was observed at a frequency of 0.27 in these animals. No significant difference was observed on androstenone measured at 100 kg or at slaughter between C/C and C/T animals.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
As pointed out by de Koning et al. (1999) and Bidanel et al. (2001a), the use of both LC and HFS models allows different a priori assumptions about QTL genotypes in founder populations to be compared. The LC model assumes that alternative QTL alleles are fixed in founder populations. It is a very powerful model when this corresponds to the true state of nature, but its robustness to departures from this ideal situation is limited (Alfonso and Haley, 1998). The HFS model does not make any assumption about the number and the frequency of QTL alleles in founder populations. It may thus be considered as a more general model, which will be able to detect QTL segregating in founder populations.
In any case, the two models gave very similar results on SSC 7 and 14 and relatively consistent ones on SSC 3 and 4. This similarity of LC and HFS results for the most important QTL tends to indicate that different alleles are likely to be fixed in these founder populations. This hypothesis is strengthened by the absence of significant family x QTL interaction in the F2 population and supports the adequacy of the regression approach (LC model) for the analysis of such LC data. Nevertheless, the presence of several alleles segregating in the parental populations cannot be totally excluded, as the experiment may somewhat lack power to test such an interaction.
Less consistent results between LC and HFS models were obtained for the other SSC. Suggestive or significant QTL were detected on SSC 9, 10, and 11, but only with the LC model, whereas on SSC 2, 6, and 18 QTL were only detected with HFS model. These results may in both cases correspond to false positives. Yet, the lack of significant results with HFS model may also be due to a loss of power associated with the much larger number of parameters that have to be estimated. Concerning QTL detected only with the HFS model, they might also be loci with similar allele frequencies in founder populations.
No statistical evidence of linked QTL could be obtained from two-QTL analyses, even though some likelihood profiles might suggest their existence. As emphasized by Knott et al. (1998), separating two closely linked QTL is difficult due to lack of recombination, and the current experiment might lack adequate power to detect linked QTL.
No chromosomal region showed any evidence of imprinting effects. Similarly, only suggestive evidence of imprinting could be found by Quintanilla et al. (2002) and Milan et al. (2002) in the same experiment. Conversely, several chromosomal regions displaying significant imprinting effects were evidenced in the Dutch QTL experiment (de Koning et al., 2000, 2001; Hirooka et al., 2001).
QTL Mapping
This study is the first attempt to identify chromosomal regions contributing to the variation of fat androstenone levels by QTL analysis. Yet, a major gene affecting fat androstenone level and bulbo-urethral gland size had previously been detected by Fouilloux et al. (1997) using segregation analysis. The results obtained here clearly show that several genome regions significantly contribute to the genetic variation of fat androstenone level. A total of six genomewide significant QTL were detected on SSC 3, 4, 7, 14, and at the end of the long arm of SSC 6 and 9.
The effects of QTL alleles are generally consistent with breed differences, with MS alleles leading to higher fat androstenone levels. Exceptions were SSC 7 and, to some extent, SSC 10 and 11, where MS alleles are associated with low androstenone levels, and SSC 6, where effects differ according to families. The transgressive variation observed on SSC 7 has been observed for most QTL detected in the SLA region (Bidanel and Rothschild, 2002). Its origin is not clearly understood but might be related to the important role of this region on many immunological pathways.
Most QTL have essentially additive effects, except on SSC 4 and 13, where MS alleles are, respectively, recessive and dominant with respect to LW alleles, and on SSC 10, where the QTL is mainly overdominant. The SSC 4 QTL thus is the only one whose effects are similar to those estimated by Fouilloux et al. (1997) using segregation analysis.
The largest effects observed in this study were obtained for the SLA region on SSC 7. Many QTL affecting growth, carcass composition, reproduction, and meat quality traits have been detected in this region (Bidanel and Rothschild, 2002). In the same population of animals, QTL associated with the development of male reproductive organs (Bidanel et al., 2001b) and of fat tissues (Bidanel et al., 2001a; Milan et al., 2002) have been evidenced in this same chromosomal region. It might correspond to a single QTL affecting the overall rate of steroid biosynthesis, leading to variations in the accumulation of androstenone in fat tissues and to differences in androgens/estrogens production, with pleiotropic effects on male sexual development and fatness deposition. This hypothesis is supported by the significant positive correlations observed between male sexual development traits and fat androstenone levels, similar to previous observations (e.g., Bonneau and Russeil, 1985).
A similar hypothesis can be put forward for the fat androstenone QTL detected on SSC 3, 4, and 10, as QTL affecting the development of male sexual organs (and for SSC 4, fatness deposition) have been evidenced in the same chromosomal regions (Bidanel et al., 2001a, b; Milan et al., 2002). The present experimental design is not appropriate to compare and test the hypotheses of linked QTL vs a single pleiotropic QTL in each chromosomal region. This might be carried out using fine mapping designs that take advantage of the accumulation of recombinant events over generations, such as advance-intercross lines (see the review of Darvasi, 1998).
In the case of the QTL found on SSC 3, the hypothesis that the effect on androstenone is the result of an effect on sexual maturity is further supported by the observation that the QTL effect is highly significant in young animals but is no longer significant at slaughter, when all animals are sexually mature, even the late maturing ones. Conversely, QTL found on SSC 4, 7, and 14 have suggestive or significant effects at all ages. Moreover, the correlations between androstenone and male sexual development traits account for only 23%, or less, of the total androstenone variance. Therefore, our results are also compatible with a specific effect of the QTL on 16-androstenes and/or androstenone production rates. This specific effect might proceed from two different mechanisms. A first hypothesis is related to the enzyme system that controls the balance between the 16-androstenes and androgens/estrogens pathways. Indeed the conversion of C21 steroids to 16-androstenes is catalyzed by cytochrome P450C17 and cytochrome b5 (Nakajin et al., 1985; Meadus et al., 1993) and the testicular levels of the low molecular weight form of cytochrome b5 are positively correlated to fat androstenone levels and 16-androstene steroid synthesis rates in vitro (Davis and Squires, 1999). The cytochrome P450C17 locus is located on SSC 14 and might be a positional candidate locus for the QTL detected on that SSC, even if their most likely positions do not exactly coincide. The cytochrome b5 locus is located on SSC 1 and is consequently not a positional candidate, but cytochrome b5 regulating genes might be worth investigating. A second hypothesis would be an effect of the specific 5-reductase (SRD5A2) enzyme that catalyzes the final step of androstenone formation and is different from the 5-reductase involved in the androgen biosynthesis pathways (Cooke et al., 1997). This gene is localized on human HSA 2p23, which is likely to correspond to SSC 3q24-27 region (Goureau et al., 1996), outside the location interval of the QTL detected on SSC 3. Direct evidence for genetic variation at the SRD5A2 gene was not detected as a QTL peak near the region, but its involvement with a possible regulatory gene might be worth investigating.
Positional candidates were first investigated for the QTL with the largest effects, the SSC 7 QTL. Both CYP11a and CYP21 loci appeared as positional candidates. The CYP21 locus appeared as particularly interesting, as it had previously been mapped in the SLA complex (Geffrotin et al., 1990), close to the most likely location of the QTL, between markers LRA1 and S0102. This enzyme induces the monooxygenation of progesterone, producing deoxycorticosterone, which is a precursor of androstenone. Yet, neither missense mutation nor polymorphism in the splicing sequences could be identified when comparing the whole sequence of CYP21 in homozygous LW and MS founder pigs. However, the existence of a mutation in the regulatory region of the gene, which would modify its pattern of expression and induce the observed QTL effect, cannot be excluded.
The CYP11a locus was proposed as a candidate gene for fat androstenone levels by Greger (2000) in a population study. Significant differences between C/C, C/T, and T/T animals were reported, the presence of a T-allele being associated with reduced fat androstenone levels. Yet, the fine mapping of CYP11a on IMpRH radiation hybrid panel led to a position that differs by more than 20 cM from the most likely position of the QTL. Moreover, the polymorphism of the C/T-155 mutation in founder pigs and in F1 males was not consistent with the observed QTL effects. A single heterozygous C/T F1 male was observed, but contrary to Greger (2000), the T-allele was associated with an increased fat androstenone level. Similarly, no difference was found between C/C and C/T genotypes at the C/T-155 mutation in the Large White x (Landrace x Large White) population investigated. These results indicate that, at least in the two populations investigated, the mutation described by Greger (2000) is not responsible for the observed variation of fat androstenone levels. Yet, it could not be determined whether the C- and T-associated haplotypes were similar or not to those found in the population studied by Greger (2000). Hence, though rather unlikely, the hypothesis of a second mutation in the two populations investigated that inverts the effect observed by Greger (2000) could not be totally excluded.
|
Implications |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
|
Footnotes |
|---|
2 On leave from: Departamento de Producción Agraria, Universidad Pública de Navarra, Pamplona, Spain.
Received for publication April 12, 2002. Accepted for publication September 18, 2002.
|
Literature Cited |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
-androstenone levels. Anim. Reprod. Sci. 15:259–263.
-androstenone content in the young boar. Anim. Reprod. Sci. 15:241–258.
-androst-16-en-3-one in adipose tissue of pigs with a rapid microtitre plate enzyme-immunoassay (MTE). Archiv für Lebensmittelhygiene 39:87–90.
This article has been cited by other articles:
|
|
 |
|
  G. Chen, E. Bourneuf, S. Marklund, G. Zamaratskaia, A. Madej, and K. Lundstrom Gene expression of 3{beta}-hydroxysteroid dehydrogenase and 17{beta}-hydroxysteroid dehydrogenase in relation to androstenone, testosterone, and estrone sulphate in gonadally intact male and castrated pigs J Anim Sci, October 1, 2007; 85(10): 2457 - 2463. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. I. Nicolau-Solano, J. D. McGivan, F. M. Whittington, G. J. Nieuwhof, J. D. Wood, and O. Doran Relationship between the expression of hepatic but not testicular 3{beta}-hydroxysteroid dehydrogenase with androstenone deposition in pig adipose tissue J Anim Sci, October 1, 2006; 84(10): 2809 - 2817. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P A Sinclair, W J Gilmore, Z Lin, Y Lou, and E J Squires Molecular cloning and regulation of porcine SULT2A1: relationship between SULT2A1 expression and sulfoconjugation of androstenone. J. Mol. Endocrinol., April 1, 2006; 36(2): 301 - 311. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Demeure, M. P. Sanchez, J. Riquet, N. Iannuccelli, J. Demars, K. Feve, L. Kernaleguen, J. Gogue, Y. Billon, J. C. Caritez, et al. Exclusion of the swine leukocyte antigens as candidate region and reduction of the position interval for the Sus scrofa chromosome 7 QTL affecting growth and fatness J Anim Sci, September 1, 2005; 83(9): 1979 - 1987. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Varona, O. Vidal, R. Quintanilla, M. Gil, A. Sanchez, J. M. Folch, M. Hortos, M. A. Rius, M. Amills, and J. L. Noguera Bayesian analysis of quantitative trait loci for boar taint in a Landrace outbred population J Anim Sci, February 1, 2005; 83(2): 301 - 307. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Nii, T. Hayashi, S. Mikawa, F. Tani, A. Niki, N. Mori, Y. Uchida, N. Fujishima-Kanaya, M. Komatsu, and T. Awata Quantitative trait loci mapping for meat quality and muscle fiber traits in a Japanese wild boar x Large White intercross J Anim Sci, February 1, 2005; 83(2): 308 - 315. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
