| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Animal Science, Texas A&M University, College Station 77843 and
and
Department of Genetics, Faculty of Veterinary Medicine, University of Liège (B43), 4000-Liège, Belgium
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
Key Words: Beef Cattle • Carcass Composition • Growth • Mapping • Quantitative Trait Loci
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
). Phenotypic selection for growth has been highly effective in beef cattle due to the ease of measurement and intermediate to high heritabilities (Koots et al., 1994a). However, there are also moderate to high genetic correlations between growth traits at different ages (Koots et al., 1994b), and selection to increase postnatal growth usually results in an increase in birth weight and thus calving difficulty. Selection based on identified genes or QTL that influence weight at weaning and at maturity, but not during the prenatal period, will facilitate the uncoupling of these strong positive genetic correlations and will allow an increase in postnatal growth while maintaining constant birth weight. Crossbreeding between Bos taurus and Bos indicus cattle has been widely practiced in subtropical regions due to the benefits of heterosis and breed complementarity for reproduction, growth, and carcass traits. If QTL influencing variation in carcass quality between these subspecies can be identified and employed for marker-assisted selection within the crossbreeding scheme, superior breeding and market animals could be identified at early stages of production. This would allow improved annual selection response and the implementation of more efficient management practices to match production to final retail product targets.
The purpose of this study was to detect QTL influencing growth and fatness traits segregating between and within the Angus and Brahman cattle breeds by applying two approaches based on line-cross and random infinite alleles models.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
Among the traits analyzed in this study were birth weight (BWT); weaning weight (WWT); yearling weight (YWT); slaughter weight (SWT); hot carcass weight (HCWT, a measure of trimmed final carcass weight at approximately 20 mo of age); adjusted subcutaneous fat thickness between the 12th and 13th ribs (ABF); and percentage of kidney, pelvic, and heart fat relative to carcass weight (KPH). The two carcass fatness traits were evaluated according to United States Department of Agriculture specifications (USDA, 1989).
Four hundred seventeen genetic markers, mainly microsatellites, were scored for the construction of linkage maps using CRI-MAP version 2.4 (Green et al., 1990) following the protocols of Beever et al. (1996). When multiple markers (usually derived from a single bacterial artificial chromosome clone) were linked with no recombinants separating adjacent markers, data were combined to produce a haplotype and the more informative haplotyped marker was used, thereby reducing the total to 357 markers for the QTL analysis.
QTL-Mapping Approaches
Two interval-mapping approaches were applied for the detection of QTL segregating between or within breeds: multiple regression interval mapping under the line-cross model (LC) (Haley and Knott, 1992), and approximate restricted maximum likelihood (REML) mapping under the random model (RA) (Grignola et al., 1996a).
The Line-Cross Model.
Under the LC model, alternate QTL alleles are assumed to be fixed or highly skewed in frequency between the breeds (Haley et al., 1994). For each progeny, the maternal and paternal inheritance was determined for each marker allele so as to assign the breed of origin of each allele using the three-generation pedigree, parental breed information, and identity by descent data from the CHROMPIC option of CRI-MAP. Thus, noninformative genotypes concerning the breed of origin of both parental alleles were deleted. The LC model at a putative QTL position is
 |
where Y is a vector of phenotypes; X is a design matrix; b is a vector of fixed and covariate effects; a is the additive QTL effect, modeled as half of the difference between Angus and Brahman breed homozygotes (2a = AA - BB); d is the dominance effect, modeled as the difference between the average of Angus and Brahman heterozygotes (AB and BA) and the homozygote midpoint; and xa and xd are vectors containing functions of genotype probabilities for each animal at the chromosomal position of the putative QTL conditional on flanking maker genotypes. The term e is a vector of uncorrelated residuals with constant variance. Fixed effects in the model were included for year season of birth, gender, cross type (two double reciprocal backcrosses and F2), and families nested within cross type. Covariates were weaning age for WWT, yearling age for YWT, and days on feed and age at slaughter for postslaughter measures. As the effect of recipient dams was significant for BWT and WWT (P < 0.05 for the 
likelihood ratio test comparing models includingand omitting the random recipient effect), observed phenotypes for these traits were preadjusted for the random effect using the predicted values produced by MTDFREML (Boldman et al., 1995). Genotype probabilities for QTL conditional on the flanking marker genotypes and QTL position were calculated, according to the cross type of each animal. For example, the element of xa was P(QQ) - P(qq), P(QQ), or -P(qq) for the F2, BCA, and BCB cross types, respectively, where Q and q represent alternate breed QTL alleles for Angus and Brahman, respectively. The F-ratios testing the null hypothesis of no QTL were obtained at 1-cM increments along each test chromosome.
The Random Model.
The RA model of Grignola et al. (1996a) is based on modeling covariances among relatives at individual marked QTL and assigning random effects to the QTL alleles within the (grand)parents of each family. This model incorporates variance components due to the polygenic and QTL allele effects, and is
 |
where Y is a vector of phenotypes, X is a design matrix, ß is a vector of fixed and covariate effects, Z is an incidence matrix relating records to individuals, u is a vector of residual additive (polygenic) effects, T is an incidence matrix relating n individuals to 2n alleles at each QTL, v is a vector of allelic effects of the QTL, and e is a vector of uncorrelated residuals with constant variance. Phenotypes, fixed effects, and covariates in this model were the same (except for the effect of families nested within cross type) in the LC model. Complete pedigree information was used to specify Var(u) and Var(v). Four unknown parameters were fitted in the model; heritability (h2), the fraction of additive genetic variance explained by QTL allelic variance (v2), residual variance 
, and QTL map position (p). The REML solution is obtained at the maximum of the restricted likelihood at which point a likelihood-ratio test statistic (LRT) of the form -2 ln(L0/L1) can be constructed to test hypotheses concerning v2. A test for the presence of a QTL is constructed with L0 the maximum value of the likelihood under the null hypothesis of no QTL allelic variance 
, and L1 the maximum value of the likelihood under the alternative hypothesis 
. Since the capacity for fitting linked markers on the test chromosome was limited to 10 in the MQAREML program implemented by Grignola et al. (1996a; Hoeschele et al., 1998), where the number of chromosomal markers exceeded 10, the LRT profile was plotted separately for each 10-marker group. The connection region for the two overlapping marker groups was determined such that the LRT values obtained from the two separate analyses were the same for at least 10 cM.
Significance Thresholds
Significance threshold values for QTL detection were based on single-trait analysis and genomewide "suggestive" and "significant" linkage levels as defined by Lander and Kruglyak (1995). In the LC model, permutation tests were performed with 10,000 replicates to empirically determine P-values at the chromosomewide (CW) significance level (Churchill and Doerge, 1994). Permutation of the phenotypes, fixed factors, and covariates to marker genotypes were restricted to within each of the three cross types. The P-value for a genomewide (GW) significance level was then obtained using the Bonferroni correction:
 |
where r is the proportion of total genome length attributed to the chromosome (de Koning et al., 2001). When the genomewide P-value < 0.05, it was concluded that the QTL was detected at a significant level of evidence for linkage.
In the RA model, 1,000 simulations were performed to obtain approximate thresholds for the CW level of evidence for linkage. Because the REML analyses are computer intensive (Grignola et al., 1996a), chromosomes were classified into six groups—high (4–6 cM), moderate (7–9 cM), and low (>10 cM) average intermarker interval size, and either large (>100 cM) or small chromosome length—and a test chromosome was randomly chosen from each group. Phenotypes of parents (F0) were randomly generated by using Gaussian distributions for breeding values and residuals using the additive and residual variances that were estimated under the no QTL model by the REML analyses. Phenotypes of progeny (F1 and F2) were determined from the breeding values of their parents, random Mendelian, and residual effects. Marker alleles for each parent were generated independently of the alleles at flanking markers (e.g., assuming linkage equilibrium). Because the allele distributions between Angus and Brahman were significantly different for most markers (data not shown), parental alleles were generated according to the different breed allele frequencies. Each of the parental haplotypes and permissible recombinants were randomly transmitted to offspring with probabilities determined by recombination rates. For chromosomes with more than 10 markers requiring more than two REML analyses, the largest LRT value was chosen across the chromosome segments. The simulation results indicated that the empirical chi-squared distributions for the LRT statistics under the condition of no QTL were close to 1 or 2 df or in between 1 and 2 df across different chromosomes or traits. We, however, decided to choose CW P-values from chi-squared distributions with 2 df to apply stringent thresholds as did Grignola et al. (1996b).
Two QTL Analyses
A one- vs. two-QTL test was performed for some QTL that were detected with a suggestive level of evidence and with test statistic profiles that were either broad curves or bimodal under the LC or RA model. For the LC model, a two-dimensional grid search on QTL positions was performed using 1-cM increments to simultaneously fit two QTL in the model. The maximum F-statistic value was obtained by dividing the difference between the residual sum of squares (RSS) from the best one-QTL and two-QTL models by two times (two parameters for the second QTL) the residual mean squares from the best two-QTL model. To obtain threshold values for the test statistic of one vs. two QTL, the data may first be preadjusted for the effect of the QTL detected to have the greatest effect in the two-QTL analysis, permuting the adjusted sample, and then searching the chromosome for the position corresponding to the maximum F-value for the second QTL. The empirical distribution of the F-statistics obtained in this manner is very similar to that obtained from a test of one vs. no QTL (J. C. M. Dekkers, personal communication). Therefore, we used the empirical distributions obtained for the test of one vs. no QTL models for this purpose. In the RA model, the most likely positions (d1, d2) for two-QTL models were obtained by a cyclic optimization algorithm (Zhang et al., 1998). A LRT value for the best two-QTL model vs. the best one QTL model (-2 ln(L1/L(d1, d2)) was compared with a chi-squared test statistic with 1 df (Zhang et al., 1998). If the LRT value was greater than the threshold of the test statistic, it was concluded that two QTL resided on the test chromosome. The significance thresholds were determined at 5% CW level.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
contains the summary statistics for the observed growth and carcass fatness traits. There were small differences in mean BWT or WWT among the 3/4Angus (BCA) and 3/4Brahman (BCB) progeny. However, 3/4Angus progeny produced heavier and fatter carcasses (SWT, HCWT, ABF, and KPH) (Kim and Taylor, 2001). Heritability estimates obtained with simultaneous adjustment for systematic environmental factors in MQAREML were intermediate (0.41 for ABF) to high (0.55 for WWT and SWT). The final map comprised 29 bovine autosomes (BTA), and the sex-average map spanned 2,642.5 Kosambi cM with an average intermarker distance of 8.1 ± 7.1 cM. Figure 1 shows the estimated map length, number of markers, and multipoint percentage of informative meioses (according to the protocol of Knott et al. [1998]) across 29 bovine chromosomes (78% average) under the line-cross (LC) model. Markers were also tested for segregation distortion to determine whether an excess of one breed’s alleles or of heterozygotes existed in any genomic region (Knott et al., 1998). However, no evidence of segregation distortion was found at a P < 0.05 GW significance level (data not shown).
|
|
and 3 show the most likely positions, F-ratios, or LRT scores and estimated effects of detected QTL under the LC and RA models, respectively. Test statistic profiles on BTA 1, 2, and 5 are presented in Figure 2. A total of 35 putative QTL were found on 19 bovine autosomes; 5 QTL with significant (1 and 4 in the LC and RA models, respectively) and 30 with suggestive (19 and 11 in the LC and RA models, respectively) evidence for linkage. The number of QTL detected for each trait was six for BWT (two significant), two for WWT, nine for YWT (two significant), five for SWT, eight for HCWT (one significant), three for ABF, and two for KPH.
|
|
|
). Also, all the QTL detected in the LC model explained small portions of the phenotypic variance among the traits (largest QTL effects were for YWT on BTA 1, for SWT and HCWT on BTA 2, and for KPH on BTA 15, which explained less than 3.5% of the phenotypic variance). The proportion (v2) of additive genetic variance due to QTL allelic variance under the RA model ranged from 8 to 41%, with an average of 18 ± 11% (Table 3). Because the average number of progeny per family was small, sampling effects may have led to these effects being overestimated (Grignola et al., 1996b).
QTL Analyses for Growth Traits
Under the RA model, two QTL affecting BWT were detected in the proximal region of BTA 6 and the distal region of BTA 2 with significant evidence for linkage (Table 3 and Figure 2). The BWT QTL detected on BTA 5 with highly suggestive evidence (P = 0.12 at GW level) for linkage under the LC model had both additive and dominance effects. For this QTL, the average Brahman allele conferred a birth weight 1.21 kg higher than the average Angus allele (Table 2). A QTL affecting YWT was detected on BTA 1 with significant evidence for linkage under the LC model and the most likely position was at 68 cM. Two QTL for SWT and HCWT were also found at, or close to, this position with suggestive evidence for linkage (Figure 2).
The LC analyses identified multiple QTL for postnatal growth traits on BTA 2. One group of QTL for YWT, SWT, and HCWT were positioned in the proximal region of the chromosome with highly suggestive evidence for linkage (P = 0.11 and 0.15 at GW levels for SWT and HCWT, respectively). Another QTL for SWT was detected in the interstitial region (72 cM) of the chromosome with suggestive evidence for linkage (in Table 2 and Figure 2). The two-QTL analyses supported the evidence for two QTL at positions similar to the locations of the QTL detected under the one-QTL model (9 cM and 72 cM for SWT and HCWT, and 18 cM and 83 cM for YWT with CW P = 0.13, 0.19, and 0.15 levels, respectively). Because the relative magnitudes of the gene effects for the QTL affecting YWT, SWT, and HCWT in the proximal region of BTA 2 were all very similar, these QTL are likely to be a single pleiotropic gene. The REML analysis also found one YWT QTL in the centromeric region of BTA 2. The two YWT QTL that were detected under the RA and LC models in the proximal region of BTA 2 were positioned at 2 cM and 18 cM, respectively, and the QTL detected in the LC model was completely dominant (Table 2).
The LRT profiles from the REML analysis of BTA 5 for YWT provided evidence for multiple QTL. The QTL positioned at 79 cM had significant evidence for linkage, whereas the QTL at 50 cM had highly suggestive evidence for linkage (P = 0.09 at GW level in Table 3). The two-QTL analysis for YWT also identified two QTL at 50 cM and 81 cM with P = 0.03 CW significance. The location of the QTL at 50 cM coincided with that for the QTL for BWT detected under the LC model (Figure 2). Interestingly, for four out of the five QTL detected for YWT under the LC model, Brahman alleles conferred increased weight. All five QTL were dominant with only the QTL on BTA 25 having reduced YWT associated with the heterozygote (Table 2). A QTL with pleiotropic effects on SWT and HCWT was located at 14 cM on BTA 23 under the RA model with significant and highly suggestive evidence for linkage (P = 0.05 and 0.13 at GW levels), respectively (Table 3).
QTL Analyses for Carcass Fatness Traits
Fatness traits ABF and KPH are indicators of external and internal carcass fat, respectively. Under the LC model, only two QTL influencing ABF were detected (on BTA 1 and BTA 19) with suggestive evidence for linkage. The REML analyses also found a QTL for ABF on BTA 1; however, its most likely location was in the telomeric region of the chromosome, distinct from the location (at 1 cM) of the QTL detected under the LC model (Tables 2 and 3). Two QTL for KPH on BTA 2 and BTA 15 were detected with suggestive and highly suggestive (P = 0.09 at GW level) evidence for linkage under the RA and LC models, respectively. There were no pleiotropic QTL detected with effects on both ABF and KPH.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
; Knott et al., 1998; de Koning et al., 1999). However, if alternate (parental breed of origin) alleles are not highly skewed in frequency in the parent breeds, which is an expectation in outbred livestock populations that have not significantly evolutionarily diverged, the power of the LC model for QTL detection is substantially decreased (Alfonso and Haley, 1998). Half-sib models, which are designed to detect the substitution effect of alternate alleles at heterozygous QTL, usually within a sire, have been extensively applied in dairy cattle populations (Georges et al., 1995; Spelman et al., 1996; Zhang et al., 1998) and are insensitive to the allele frequency distribution between parental breeds (de Koning et al., 1999). However, half-sib models do not allow the partitioning of the QTL allele substitution effect into discrete additive and dominance effects, which is one of the main goals of QTL mapping in livestock populations. Also, this model has less detection power than the LC model, especially for QTL with primarily dominant gene action (de Koning et al., 1999; 2001). In this study, we applied a general approach to QTL mapping under a random model (Grignola et al., 1996a), in which random QTL allele effects are predicted using a covariance matrix for probability of identity by descent that is conditional on the observed marker data.
Only a few of the QTL that were detected with suggestive evidence for linkage under the LC model had primarily additive effects (i.e., five QTL for BWT on BTA 3, for HCWT on BTA 13 and BTA 29, for ABF on BTA 19, and for KPH on BTA 15), whereas 15 QTL were found under the RA model, in which QTL can be detected with large allele substitution effects or additive effects if alternate alleles are equally distributed such that 
= a + d(q - p) = a (Falconer and Mackay, 1996). This suggests that alleles for QTL with primary additive effects influencing growth and carcass fatness are segregating within both the Angus and Brahman parental populations.
In contrast, 15 of the 21 QTL detected under the LC model had a dominance mode of gene action (with overdominance for the QTL for YWT, SWT, and HCWT on BTA 1; for SWT on BTA 2; and for HCWT on BTA 22 in Table 2). Also, none of the QTL detected under the LC model were found by the REML analyses (except the QTL for YWT on BTA 5), which must be the case if the QTL had large additive effects (de Koning et al., 1999; 2001). Further, the estimates of the proportion of phenotypic variance due to the QTL under the LC model were not large (the largest QTL explains only 3.3% of the phenotypic variance [Table 2]) partly because the dominance effect contributes less to the QTL genetic variance than does the additive effect (VQTL = Vadditive + Vdominance = 0.5a2 + 0.25d2 under the assumption of equal breed QTL allele frequencies (Falconer and Mackay, 1996)). Heterosis for preweaning traits in F1 calves and/or F1 cows and for postweaning and carcass weight has been exploited in subspecific crosses of Bos taurus and Bos indicus (Franke, 1980; Baker et al., 1984; Comerford et al., 1988). This heterosis may be due in large part to the dominant QTL influencing the growth traits that were detected in this study.
For several decades, selection on preweaning and postweaning growth and on growth-related traits, such as average daily gain, feed conversion, and body size, has been implemented to improve beef productivity in both Bos taurus and Bos indicus breeds in the United States (Koch et al., 1974; Buchanan et al., 1982; Vargas et al., 2000). Therefore, it is likely that favorable alleles for growth have increased in frequency at QTL within both Bos taurus (Angus) and Bos indicus (Brahman) breeds, and thus the distributions of allele frequencies at these QTL have become similar between both populations. Consequently, it is likely that QTL with primarily additive effects will not be identified under the LC model, because the alternate breed alleles that were randomly sampled in families have similar magnitudes of additive effects (Alfonso and Haley, 1998).
The QTL for YWT, SWT, and HCWT on BTA 1 detected under the LC model had an overdominant mode of gene action. However, the genetic basis for these effects is not clear. There may be a single pleiotropic underlying gene, or several closely linked genes with identical directional dominance but alternating breed additive effects, or even epistatic effects among these closely linked QTL as has been reported for plants (Eshed and Zamir, 1995; Cockerham and Zeng, 1996; Hollick and Chandler, 1998). The F-statistic profiles for the QTL were very broad, which suggests that more than one closely linked QTL affecting postweaning growth may reside on this chromosome (Figure 2). However, the two-QTL analyses for YWT demonstrated only limited evidence for two QTL (at 65 cM and 134 cM with CW P = 0.45 significance level). To further test for the presence of multiple QTL, 10 half-sib families from either an F1 sire or F1 dam with at least 30 progeny (average 52, maximum 87) were chosen such that each F1 parent had both Angus and Brahman, and/or F1 mates, and the LC analyses were performed for each family. In one group of two families, the most likely positions of the YWT QTL were at 62 and 67 cM (P = 0.04 and 0.24 at CW levels), respectively, very close to the most likely position (66 cM) of the QTL detected in the overall analysis. The signs of these QTL effects were the same between the families (-2.9 and -6.8 kg for additive [a] and 27.5 and 14.7 kg for dominance [d] effects). In another group of two families, the most likely positions were at 83 and 84 cM (P = 0.35 and 0.10 at CW levels), respectively (3.4 and -34.7 kg for additive [a] and 41.7 and 6.5 kg for dominance [d] effects). In another group of two families, the most likely positions were at 47 and 53 cM (P = 0.01 and 0.03 at CW levels), respectively. However, the signs of the QTL effects differed between the two families (46.4 and -31.7 kg for additive [a] and 59.2 and -60.0 kg for dominance [d] effects). This suggests that either several QTL with distinct modes of gene action lie in the 47- to 53-cM region of BTA 1, or there is a single QTL with several alleles segregating among the breeds, which did not allow the QTL to be detected by the overall analysis under the LC model (de Koning et al., 1999; 2001). Clearly, there is strong evidence for two distinct QTL on BTA 1 that influence YWT.
No pleiotropic QTL were found affecting both ABF and KPH. This result is not surprising, considering the other reports where very few pleiotropic, or closely linked, QTL for internal and external fatness traits were detected in whole-genome scans in agricultural animals (Andersson et al., 1994; de Koning et al., 1999; Stone et al., 1999). This may suggest the involvement of different genes of large effect affecting internal and external fat depots. One example of such a gene is heart fatty acid-binding protein (H-FABP), which is located in a region of swine chromosome 6, where two QTL for intramuscular fat and backfat thickness were detected in an experimental pig population (Gerbens et al., 2000). Gerbens et al. (2000) found only an association between H-FABP and intramuscular fat.
There are some reports concerning QTL affecting growth and carcass fatness in cattle. Grosz and MacNeil (2001) found a QTL for birth weight on BTA 2 with significant evidence for linkage in a backcross population of Hereford and Bos taurus composite breeds. This study detected a QTL affecting BWT in the identical telomeric region of BTA 2 with significant evidence for linkage under the RA model. Casas et al. (2000) found a BWT QTL on BTA 6 with significance evidence of linkage in two half-sib families with 1/2 Belgian Blue and 1/2 Piedmontese F1 sires, respectively. Li et al. (2002) found associations between birth weight and haplotype segments on BTA 5 in a commercial population of Bos taurus breeds. We found BWT QTL on BTA 6 and BTA 5 under the RA and LC models with significant and highly suggestive evidence for linkage, respectively. Beever et al. (1997) found QTL for adjusted WWT on BTA 3 and 12 and for yearling weight on BTA 23 in nine half-sib families sired by Simmental, Gelbvieh, or South Devon (Bos taurus) breeds. They reported that the QTL affecting WWT on BTA 12 was associated with the bovine B erythrocyte antigen (EAB). We detected a QTL for WWT on BTA 12 that was flanked by EAB (Table 3). Elo et al. (1999) reported a QTL for daughter’s live weight during the first four lactations, which was flanked by markers BM1258 and BoLA-DRBP1 on BTA 23 in Finnish Ayrshire dairy cattle. We detected QTL affecting SWT and HCWT with suggestive and significant evidence for linkage, respectively at the location of RM33, which is adjacent to the two markers used by Elo et al. (1999) (Table 3).
The development and application of QTL fine mapping methodologies has recently been proceeding in experimental and commercial livestock populations (Marklund et al., 1999; Meuwissen et al., 2002; Blott et al., 2003). Fine mapping will refine the size of the chromosomal regions harboring the QTL detected in this experiment, which will enable breeding schemes in composite cattle to employ marker-assisted selection and will allow positional candidate gene analyses to proceed with high levels of accuracy and precision.
|
Implications |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
|
Footnotes |
|---|
3 Present address: Department of Animal Science, 233 Kildee Hall, Iowa State University, Ames 50011-3150.
4 Present address: Department of Animal Sciences, S135 ASRC, University of Missouri, 920 East Campus Drive, Columbia 65211-5300.
2 Correspondence—phone: 515-294-2721; fax: 515-294-9150; E-mail: kimjj{at}iastate.edu.
Received for publication April 3, 2003. Accepted for publication May 21, 2003.
|
Literature Cited |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Gutierrez-Gil, J. L. Williams, D. Homer, D. Burton, C. S. Haley, and P. Wiener Search for quantitative trait loci affecting growth and carcass traits in a cross population of beef and dairy cattle J Anim Sci, January 1, 2009; 87(1): 24 - 36. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Garrett, G. Rincon, J. F. Medrano, M. A. Elzo, G. A. Silver, and M. G. Thomas Promoter region of the bovine growth hormone receptor gene: Single nucleotide polymorphism discovery in cattle and association with performance in Brangus bulls J Anim Sci, December 1, 2008; 86(12): 3315 - 3323. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Abe, J. Saburi, H. Hasebe, T. Nakagawa, T. Kawamura, K. Saito, T. Nade, S. Misumi, T. Okumura, K. Kuchida, et al. Bovine quantitative trait loci analysis for growth, carcass, and meat quality traits in an F2 population from a cross between Japanese Black and Limousin J Anim Sci, November 1, 2008; 86(11): 2821 - 2832. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Nkrumah, E. L. Sherman, C. Li, E. Marques, D. H. Crews Jr., R. Bartusiak, B. Murdoch, Z. Wang, J. A. Basarab, and S. S. Moore Primary genome scan to identify putative quantitative trait loci for feedlot growth rate, feed intake, and feed efficiency of beef cattle J Anim Sci, December 1, 2007; 85(12): 3170 - 3181. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Sellner, J. W. Kim, M. C. McClure, K. H. Taylor, R. D. Schnabel, and J. F. Taylor BOARD-INVITED REVIEW: Applications of genomic information in livestock J Anim Sci, December 1, 2007; 85(12): 3148 - 3158. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. K. Musani, N. D. Halbert, D. T. Redden, D. B. Allison, and J. N. Derr Marker Genotypes and Population Admixture and Their Association With Body Weight, Height and Relative Body Mass in United States Federal Bison Herds Genetics, October 1, 2006; 174(2): 775 - 783. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Kneeland, C. Li, J. Basarab, W. M. Snelling, B. Benkel, B. Murdoch, C. Hansen, and S. S. Moore Identification and fine mapping of quantitative trait loci for growth traits on bovine chromosomes 2, 6, 14, 19, 21, and 23 within one commercial line of Bos taurus J Anim Sci, December 1, 2004; 82(12): 3405 - 3414. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Mizoshita, T. Watanabe, H. Hayashi, C. Kubota, H. Yamakuchi, J. Todoroki, and Y. Sugimoto Quantitative trait loci analysis for growth and carcass traits in a half-sib family of purebred Japanese Black (Wagyu) cattle J Anim Sci, December 1, 2004; 82(12): 3415 - 3420. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Casas, S. D. Shackelford, J. W. Keele, M. Koohmaraie, T. P. L. Smith, and R. T. Stone Detection of quantitative trait loci for growth and carcass composition in cattle J Anim Sci, December 1, 2003; 81(12): 2976 - 2983. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
