MXPA06009037A - Marker assisted best linear unbiased predicted (ma-blup):software adaptions for practical applications for large breeding populations in farm animal species - Google Patents

Abstract

The invention provides methodologies for improved molecular genetic analysis of individual animals and animal populations. The invention includes methods and systems for identifying those animals in a population that are most likely to heritably pass on desirable traits. Provided are means for evaluating the estimated breeding values and increasing the average genetic merit for animals in a population. For each trait, the instant invention provides methods for evaluating the relative effect of one or more quantitative trait loci (QTL) and three or more molecular genetic markers for each QTL. The relationship between these various markers and the pre-selected trait and QTL is calculated, along with the contribution of other factors such as pedigree and known measures with respect to quantitative trait, and these data are used to calculate estimated breeding values for the animals in the herd and to rank the animals according to these estimated breeding values.

Description

BEST LINEAR IMPEDCIAL PREDICTION ASSISTED WITH MARKER: SOFTWARE ADAPTATIONS FOR PRACTICAL APPLICATIONS FOR LARGE FOSTER POPULATIONS IN SPECIES OF FARM ANIMALS This application claims the benefit of U.S. Provisional Application Serial No. 60 / 543,034, filed on February 9, 2004, which is incorporated herein by reference.

BACKGROUND OF THE INVENTION FIELD OF THE INVENTION The present invention relates in general to the field of improvement in the genetic importance in animal species both in individual animals and at flock levels. Among the various modalities, particularly refers to a method to improve the genetics in pig and cattle herds. More particularly, the invention provides for the analysis of multiple genetic markers as part of a herd management and breeding program.

DESCRIPTION OF THE RELATED TECHNIQUE Due to the rapidly growing field and the improvement of genomics, there is a need for means to utilize newly available genotypic information to improve the development of commercial animals and plant products. Such means should allow the rapid genetic improvement of a population in order to optimize the short-term occurrences of desirable traits in the population without risking the potential of long-term genetic improvement (for example, as documented through excessive inbreeding or intense selection pressure on a limited number of genes or quantitative trait sites (QTL) [eg, Gibson, 1994]). Such a method may need to provide means to rapidly and efficiently maximize the utility of a new understanding with respect to the function of several genes and / or combination of genes; while at the same time optimizing the use of phenotypic, genotypic, (for example, SNPs) and pedigree information. This is particularly important in traits where genotypes are difficult or expensive to measure (eg, food input or resistance / disease tolerance) traits that are measured late in life, or at the end of life (eg, quality of life). longevity or meat) or measurable only in one sex (for example, milk yield, breeding size or facilitating maternal or paternal birth). In traits such as meat quality, not only is the trait measured after the selection decisions that have already been made, but the animal has most likely been sacrificed to allow the measurement of traits and, therefore, is no longer available for selection. In these cases, marker-assisted selection (MAS) can provide extremely useful information for selection before the availability of phenotypic measurements. The present invention provides the ability to practice MAS over several QTLs in an optimal and efficient way on an industrial scale.

BRIEF DESCRIPTION OF THE INVENTION The invention now described solves the previously existing problems through the provision of a method that allows the capture of pedigree, phenotypic, and molecular genetic metrics for a population of offspring that provides the concurrent and interdependent evaluation of these factors, for each animal (or plants) and then provides a classification of the individuals in the population that enable the optimal weighting of all sources of information to achieve the desired breeding objectives. The invention now described solves the deficiencies associated with previously available technology by allowing the concurrent evaluation of one or more, two or more, or three or more molecular genetic markers, pedigree information, and optionally the metrics of the quantitative assays through the use of iteration algorithms on data (lOD) that dramatically reduce computer memory requirements and preconditioned conjugate gradient algorithms (PCCG) with blocking / common variable size preconditioner, which dramatically reduces the calculation time. The invention also provides algorithms for calculating inbreeding coefficients, in QTL. Existing software may also have the ability to incorporate marker information that is severely hampered by long calculation times and excessive computer memory requirements. By dramatically reducing computer memory requirements for solving the mixed model equations through the incorporation of IOD algorithms, several aspects of the present invention make it possible to include a virtually limited number of marked QTL and any number of features. The PCCG algorithms included in the aspects of the present invention significantly reduce the calculation time, thus allowing numbers of major markers and features to be included in the mixed model equations while adequately converged solutions are achieved over a period of time. acceptable time for breeding programs that operate on an industrial scale. The importance of being able to practice and efficiently include more markers of two has many advantages. First, while more QTL markers are included in MA-BLUP (better linear unbiased prediction aided by marker), a greater proportion of the genetic variation of the traits selected can be explained through the marker information and, therefore, the genetic progress It also accelerates. Secondly, it has also been shown that intense selection in only a few QTLs (for example, from 1 to 3 places) can accelerate the short-term genetic response, but it occurs with the cost of long-term genetic progress. In fact, it has been shown that MAS (marker-assisted selection) with only a few included sites can provide a less favorable long-term genetic response than BLUP alone (ie, no marked information is included) (Gibson, 1994). Accordingly, if the selection takes place on several markers simultaneously, as provided by the present invention, the loss of the long-term response is minimized. In several aspects of the present invention the persecuted tear (s) to be improved are selected for the presence of desirable characteristics including but not limited to: the presence or essence of the specific gene or variants of the marker or alleles, health traits , reproductive traits, meat quality traits, efficient growth traits, and any other desired phenotypic trait. Several embodiments of the present invention provide a method for increasing the genetic importance of the animal population with respect to one or more pre-selected risks. Certain aspects of this method comprise the steps of selecting one, two, three, or more molecular genetic markers of interest, for each or more of the trait sites (QTL), for the trait for which an improvement is desired. For each of the selected characteristics, either as molecular genetic marker genotypes or quantitative trait measurements, a computer readable database is provided that indicates each animal's status in the population with respect to the selected characteristics if they are available for the animal. The methods and systems of the present invention do not require that the phenotypes be available to each animal in the population (i.e. the methods and systems of the present invention are capable of handling missing terms). In addition, thanks to its multiple feature capabilities, the present invention does not require phenotypes to be available for all traits for each animal or for each marker to be effective. For example, if genotypes are available only in most recent generations in the pedigree and available for some markers or animals but not for others, the methods and systems of the present invention may still be remarkably effective. Additionally, a computer readable database that provides the pedigree for each animal in the population is also provided. A computer is then used to carry out the analysis of the best linear unbiased prediction assisted by molecular genetic marker (MA-BLUP) data in the provided databases. This analysis simultaneously produces estimates of breeding value (EBV) for each animal and for each trait using marker, pedigree, and phenotypic data, if available, on all traits simultaneously. Then there is a classification of the animals in the population where the animals are classified according to their EBV (estimated breeding value) for the combination of the EBVs of individual trait that is represented in the selection index for a given population, the which takes into account the inbreeding coefficients for the selected traits. This classification can then be used as part of an animal management or breeding plan to optimize the improvement of the average genetic importance of the population for the selected characteristics. Other embodiments of the invention provide a system for increasing an average genetic importance of the animal populations. In several aspects of this modality the system comprises a computer, one or more accessible databases, a program executable by computer, and a user interface. The databases, computers, and computer programs provided by the various aspects of this embodiment of the invention are the same as those in the methods described supra. The user interfaces considered useful for the various aspects of this embodiment of the invention are configured to be coupled with the computer to allow the user to instruct the computer to access the available databases and allow the computer program to use the processor. of the computer to generate, as an output, its estimated individual breeding value and / or one or more classifications of the animals in the population. Another embodiment for the present invention provides a method for evaluating a breeding value of the animal population or genetic significance for a preselected group of characteristics. Although the evaluation can be accomplished using one or two molecular genetic markers for each QTL, according to the various preferred aspects of this invention the features will typically include at least three molecular genetic markers. Even more preferably, the selected features will include four or more molecular genetic markers. The selected characteristics will be linked (or associated) with one or more QTLs or one or more genes of economic value. Several aspects of this embodiment of the invention provide the steps of: (a) selecting one, two, three, four or more molecular genetic markers of interest that are linked to one or more QTLs or genes; (b) providing databases comprising data for individual animals in the population, including the pedigree of the animals, and the status of the animal for each of the selected features where they are known; (c) use a program executable by computer in a computer capable of carrying out MA-BLUP to simultaneously analyze the data of the database provided to produce a classification of each animal, in the population, according to its EBV for each trait selected, taking into account the possible inbreeding; and finally, (d) evaluate the EBV's of the traits to determine the combined multi-trait EBV for the traits selected in the selection index. In this way, as provided here, MA-BLUP performs a "coupling" or simultaneous analysis to produce EBVs for each trait and each animal of the mixed model equations. These are then used in combination through MA-BLUP to provide a single value known as the "Selection Index". Other embodiments of the present invention provide useful systems for increasing a genetic importance of the animal population, wherein the system comprises the following components, (a) A computer to which the data is loaded and which is capable of operating a program of computer to produce output data, (b) At least one database accessible by computer, where the database is selected from those that provide the pedigree data for the population, databases that provide information about the place of the quantitative trait, and the molecular genetic markers (both of these markers are known to be associated with any selected quantitative trait site), (c) A computer executable program capable of simultaneously evaluating the data in all the databases providing and producing estimated breeding values of program output (EBVs) for each trait and for each individual animal in the population for each trait individually and in combination and the classification of the animals according to their respective EBVs, (d) A user interface that includes data capture and recovery systems, where the user interface is coupled to the computer and configured to allow the user to instruct the computer to access any combination of available databases and use the computer program to generate the output ratings and estimated breeding values of the individual animal. Other modalities provide for the use of any of the methods or systems described herein to evaluate the average genetic significance of an animal population for one or more selected traits. Still another embodiment of the present invention provides a method for identifying the best pairs of offspring in a defined animal population to allow for the optimal improvement of a preselected trait in the population (eg, to quickly improve the average EBV for the characteristics in the population). In accordance with this aspect of the invention, any of the methods for estimating the animal or EBVs herds for a given trait can be used as part of a method to identify those pairs of animals best suited for crossing (without exceeding a degree). by acceptable rate of inbreeding) in order to optimize the increase of the population's average breeding value or genetic significance for a pre-selected characteristic or trait. Taken together, the MA-BLUP methods and systems of the present invention provide a synergistic confluence of the elements that enable those skilled in the art to solve the equations of the mixed model which was a problem previously difficult to manipulate the pedigree (or! mpractic solve for industrial-scale populations, QTL, and molecular genetic marker data to calculate the EBV for each animal in a very large population of more than one million animals and classify each animal in that population according to its individual EBV for one or more pre-selected traits Other embodiments of the present invention provide methods for improving one or more meat quality traits, where the meat quality traits include, but are not limited to the pH of the loin and / or ham, softness, marbled color and ability to hold water. Several aspects of these modalities provide methods for classifying a plurality of pigs to identify the status of each animal with respect to one or more single nucleotide polymorphisms (SNPs) in the porcine PRKAG3 gene (the PRKAG3 gene encodes a specific muscle isoform of the regulatory kinase sub-unit of protein kinase activated by adenosine monophosphate (AMPK), the PRKAG3 strains for the gamma-3 subunit activated by protein kinase AMP). Preferably, the identified SNPs are selected from the group consisting of: an A / G at position 51, A / G at position 462, A / G at position 1011, C / T at position 1053, C / T at position 2475, A / G in position 2607, A / G in position 2906, A / G in position 2994, and C / T in position 4506, where all the numbering is in accordance with the sequence of SEC ID NO: 1. Once the animals having at least one desired allele are identified, they are selected for use as stallions / females in the breeding plan designed to produce the progeny that has an increased frequency of the desirable allele. Other modalities provide methods and / or equipment to detect the PRKAG3 SNPs described above. In addition, in several aspects of these modalities these methods and / or equipment are used as components of a general method or system that incorporates the use of the MA-BLUP analysis described herein. AND! use of the MA-BLUP integration methods and systems produce the managers of flocks of pups the necessary means to raise a herd administration and breeding plans to quickly improve the quality traits of the meat made through the pig gene PRKAG3 . The particular aspects of this modality provide the methods of classifying a population of animals to identify those animals that when they mate will probably produce progenies that exhibit improvements in at least one quality trait of desirable meat. In a particularly preferred aspect of this embodiment the quality trait of the desired meat for a higher ham or loin pH, darker in color, greater softness, more marbling, and / or an increased water holding capacity or any combination of the same. As noted several embodiments of the present invention provide useful equipment for carrying out the present invention. Several aspects of these modalities specifically provide the equipment that is useful for the detection of SNPs in the porcine PRKAG3 gene.

BRIEF DESCRIPTION OF THE DRAWINGS The drawings described are part of the present specification and are included to further demonstrate certain aspects of the present invention. The invention is best understood by reference to one or more of these drawings in combination with the detailed description of the specific embodiments presented herein. FIGURE 1: Figure 1 provides a schematic representation of the inputs and outputs of the MA-BLUP program (MA-BLUP is represented as a "black box"). FIGURE 2: Figure 2 provides a flow diagram that represents a possible algorithm to implement the MA-BLUP program described here. FIGURE 3: Figure 3 provides a flow chart that represents a possible algorithm for solving the mixed model equations (MME). This is the expanded version of the step enclosed in the rhomboid in Figure 2. FIGURE 4: The DNA sequence of the AMPK sub-unit range Sus scrofa (PRKAG3) (SEQ ID NO: 1), as provided by the number of access of Genbank AF214521. FIGURE 5: A graph describing the genotype values for SNP assays 1484004 and 148009. FIGURE 6: A graph describing breeding values for SNP assays 1484004 and 148009. FIGURE 7: DNA and amino acid sequence of the portion of the leptin receptor gene (pLEPR) and Sus scrofa containing the M69T and S731 polymorphisms. The individual nucleotide polymorphisms and the accompanying amino acid changes are shown in bold. The nucleotide sequence without the attached amino acid sequence is intronic. The sequence starts at position 311 of Genbank access AF184172, "the leptin receptor gene (LEPR) Sus scrofa, exon 4, and the partial coding sequence". The M69T polymorphism is in the 609 nucleotide position of the sequence with the accession number of Genbank AF184172.

DETAILED DESCRIPTION OF THE INVENTION The invention now described establishes a method for rapid improvement of a population of animals or plants, based on pedigree, phenotypic and / or genotypic information. In this way, by using the present invention, one skilled in the art will be able to use the newly described genetic or phenotypic information in order to produce an optimized progeny for one or more desired traits and / or increase the genetic importance of the population. for a pre-selected and / or desired feature or feature. This phenotypic / genotypic information can be obtained from a variety of sources. Such sources include, but are not limited to, marker genotypes in some or all of the animals in the breeding population, new or accumulated pedigree information and / or phenotypic trait measurement data and new biometric techniques. The present invention also provides methods, compositions, and equipment useful for improving the quality traits of meat in a population of pigs. Specifically, the present invention provides methods, compositions and equipment useful for the analysis of the state of the animals with respect to the PRKAG3 gene of swine. However, one skilled in the art will appreciate that the systems and methods described herein (including the MA-BLUP methodology) can be used effectively with all known quantitative trait sites and all known molecular genetic markers. By way of example, the investment provided here can make effective the use of polymorphisms in the melanocortin-4 receptor (MC4R) gene and the PRKAG3 gene. For the sake of simplicity the language and examples used in the present description will mainly refer to animal populations. However, in view of the present disclosure those skilled in the art will appreciate that the claimed inventions could be modified for use in plants by one skilled in the art having access to the present disclosure.

Defined terms The following definitions are provided herein for the purpose of assisting the quantitative or molecular geneticist or creator of animals skilled in the art to more readily and fully appreciate the present invention. As suggested in the definitions provided below, the definitions provided are not exclusive, unless that is indicated. Rather, they are provided as preferred definitions, provide the approach of one skilled in the art, on various illustrative embodiments of the invention. As used herein the term "acceptable degree of inbreeding" preferably means a level of inbreeding where the benefits of inbreeding outweigh any negative effects. In general, inbreeding will accumulate in an animal population as a result of intra-population selection. Typically, there is an inverse relationship between the degree of inbreeding (? F) and the degree of genetic advance (? G). The optimum (? F) is the degree to which inbreeding is allowed to accumulate in order to optimize both short-term and long-term genetic gains. Under standard practice in pigs it is typically desired that? F be maintained at less than 1% per year. The methods to approximate? F are given, infra, in the section "Illustrative Modalities". As used herein the term "allele" refers to a particular version or variant of a specified gene. As used here, the term "BLUP" (which is an acronym for the best linear unbiased prediction) refers to a statistical methodology introduced by Henderson (1959, 1963) that has become a standard in the breeding industry. animals to forecast breeding values for individual animals. With standard post-graduate training in animal husbandry techniques, BLUP can be carried out, through one skilled in the art, using any of several commercially available computer programs, which are used for the genetic evaluation of an animal. and / or herd. The most currently available programs are adapted programs designed specifically to meet the needs of the breeding company. However, some standard software packages that are publicly available can be used to perform BLUP (eg, "MTDF-REML" by Curt Van Tassell (curtvt aipl.arsusda.gov); "PEST" by Eildert Groeneveld (eg@tzv.fal.de); "DMU" by Just Jensen (lofjust@vm.uni-c.dk); "MATVEC" by Steve Kachman (www.statistics.unl.edu/faculty/steve/software/matvec/); and "BLUPF90" by Ignacy Misztal (http://nce.ads.uga.edu/~ignacy/newprograms.html)). Typical input parameters for BLUP programs include estimates of genetic and phenotypic parameters, phenotypes, pedigrees, and fixed effects. BLUP models can be more easily described in matrix annotations as follows:? = Xß + Za + e, where,? is the vector of phenotypic observations, ß is a vector of fixed effects; X is an incidence matrix related to ß relating ß to?; a is a vector of animal effects with an average of zero and a variant-variant matrix Ga; Z is an incidence matrix relating to?; ye is a vector of residual effects with the variant-co-variant matrix R. Ga can be modeled as Ga = A s2a, where A is the coefficient matrix of the additive relationship between the animals, and s2a is the genetic variant additive One of the requirements to obtain BLUP is to obtain the inverse of Ga, which can be calculated very efficiently even when extremely long data are established (Henderson, 1976, Quaas et al., 1984, Quaas, 1988).

As used herein the term "breeding plan" preferably refers to a program to improve the genetics of the flock using the information provided by the methods and systems described herein. As used herein the term "breeding value" preferably refers to the expected value of an animal as a parent. It is also a measure of the net breeding value of the animal. Half of the breeding values are transmitted to their progeny, and this portion can be referred to as the expected progeny difference (EPD) or the estimated transmitted ability (ETA). These measures of breeding values are typically expressed as a difference between the average present population or the average population at a point in the fixed time (see, Van Vleck, p.186). As the term "approximation" is used herein, when used to describe a molecular genetic marker and QTL, it preferably refers to the relative link distance or recombination probability between the site of the marker and the site responsible for the trait in a unit of Morgan (M) As used herein the term "drop loss" preferably refers to the change in the weight of a meat cut (eg, a loin cut) due to moisture loss of the absorbent packing materials over a period of time specified, especially while the meat is placed in a showcase. As the term "economic trait site" (ETL) is used herein, it preferably refers to a location on the chromosome that is linked to the "quantitative trait" that provides the economic value. As used herein, the terms "efficient growth traits" and / or "performance traits" preferably refer to a group of traits that are related to the degree of growth and / or composition of the animal's body. Examples of such traits include, but are not limited to: average daily gain, average daily food intake, feed efficiency, fat thickness in the back, muscle area in the spine, and lean percentage. As the term "estimated breeding value" (EBV) is used herein, it preferably refers to a specific numerical value for an animal that predicts its "breeding value". EBV is usually calculated using commercially available analysis programs (the output of BLUP and BLUP programs (MA-BLUP) assisted by marker are examples of EBVs). As used herein the term "gene" refers to ana DNA sequence responsible for coding instructions for making a specific protein within a cell or may also instructions for when, where, and in what abundance a protein is expressed). As used herein the term "genetic significance" refers to the value of germplasm to provide a desired trait. That is, the greater the genetic significance of an animal for a given trait, the more likely it is to provide a progeny that has the desirable trait. As used herein the term "fixed effects" preferably refers to seasonal, spatial, geographical, environmental, or administrative influences that cause a systematic effect on the phenotype or for those effects with levels that were deliberately configured through the experimenter, or the effect of a gene or allele / variant QTL that is consistent across the population being evaluated. As used here the term "half-brother" refers to a group of animals that all share a father. Specifically, the term is most frequently used as a "half-brother father" that refers to the progeny that share the same female. As used herein, the term "health traits" preferably includes any traits that improve the health of the animal and / or herd. These include, but are not limited to: the absence of abnormalities or undesirable physical defects (such as ruptures of the scrotum in pigs), improvements in the strength of legs and legs, resistance to specific diseases or disease organisms, or general resistance to pathogens. As used herein the terms "herd" and "population" refer to any group of breeding animals that have a sufficient number of animals for the effective use of the present invention. The term may be applied to animals such as pigs, cattle, goats, or any other animal that is commercially originated, including, but not limited to, poultry (such as turkeys or chickens) or any other species where it is desirable by any Another reason to analyze the multiple features in the creation of a breeding program. In addition, the term population can also be used to refer to a population of plants. As used herein the term "improved germplasm" preferably refers to a change in the genome, frequency of improved genetic markers, genes, alleles of markers or genes, or any combination of multiple markers or genes that is preferred over the other forms of the genome that exist in the population. This includes genome forms that result in improved breeding values, but for which the genotypes are unknown. The term may, depending on the context, be used to refer to the genetic makeup of any individual animal or the genetics of a herd, considered as a whole. In this way, the term "improved germplasm" covers both the introduction of a preferred trait in an individual, and an increase in the frequency of expression of a desired allele within a herd. As used herein the term "inbreeding coefficient to a QTL" preferably refers to the probability that two alleles in a QTL are identical by the offspring. These inbreeding coefficients are used in the calculation of G'1. The algorithm used to calculate the inbreeding coefficient for a QTL is a basis in the method described in Abel-Azim and Freeman (2001). As used herein, the term "informativity", when used to describe or modify the term "molecular genetic marker" preferably refers to a measurement of the marker's value as a predictive determinant for how likely a given trait and / or QTL it will be inherited by the animal's progeny. In this way, informativity is a measure of the genotypic variation present at the marker site and is determined as a measure of the frequency of heterozygosity of the marker. If a marker is informative enough and is located relatively close to the QTL location, the utility as a marker for QTL is increased. The more informative the markers surrounding a QTL, the more likely the QTL site can be defined. As used herein, the term "site" refers to a specific location on a chromosome (eg, where a gene or marker is located. "" Sites "is the plural of a site, as used herein, the term" MA-BLUP "(an acronym for BLUP assisted by marker) is a method of analysis that uses the same captures as BLUP (see above) and additionally adds the animal's marker genotype to the calculations. As with BLUP, MA-BLUP models can be more easily described in a matrix annotation as follows:? = Xß + ZK? + Zu + e where,? it is the vector of phenotypic observations; ß is a vector of fixed effects; X is an incidence matrix relating ß to?; ? is the vector of additive effects in the QTL marked with an average of zero and a variant-co-variant matrix, G ?, and u is the vector of additive effects of the remaining unlabeled QTL with an average of zero and a variant matrix - co-variant Gu (ie effects of animals, previously represented by a, and subdivided into? yu, as a = KK + u, where K is the incidence matrix that relates? a). Z are the incidence matrices that relate K? and u with?; e is a vector of residual effects with a variant-co-variant matrix R. To carry out MA-BLUP, the inverse of G? and Gu need to be calculated. The inverse of Gu can be obtained as with Ga in regular BLUP (see above). The inverse for G? it can be calculated efficiently for large groups of data where the marker genotypes can be inferred in each animal and the parental origin of the marker is known (Femando and Grossman, 1989), and in the case where the marker genotypes are not known in some animals and the parental origin of the marker is unknown (Hoeschele, 1993; van Arendonk et al., 1994; Wang et al., 1991; Wang, et al., 1995). As used herein, the terms "label" and "molecular genetic marker" (MME) preferably refer to a DNA sequence that has a specific site on a chromosome that can be measured in a laboratory. To be useful, a marker needs to have two or more alleles or variants. Common types of markers include, but are not limited to: RFLP = restriction fragment length polymorphism; SSR = simple sequence repetition (a.k.a "microsatellite" markers); and SNP = individual nucleotide polymorphism. The markers can be either direct, that is, located within a gene or site of interest, or indirect, which are closely linked to the gene or site of interest. In addition, the labels can also include sequences that either modify or do not modify the amino acid sequence of the gene.

As used herein, the term "mixed model equation" preferably refers to a model for equations that solve both random effects and fixed effects. The term random effect in the MA-BLUP context is used to denote factors that have an impact not systematic about the trait with levels that can represent a random distribution. Random effects will typically have levels that they were not deliberately organized by the experimenter (deliberately organized factors can be called fixed effects), but they were sampled from a population of possible samples rather. Linear models that incorporate both fixed effects and random effects are called mixed linear models. The best unbiased prediction linear random effects and physical effects are the solution of the following linear equations, which are called mixed model equations. y = Xb + Z1u + Zz? + e As used here, the preferred meaning for the term "marker-assisted allocation" (MAA) is the use of phenotypic or genotypic information to identify animals with estimated breeding values superiors (EBVs) and also assign to these animals a specific use designed to optimize the improvement of the genetic importance of the animal population.

As used herein, the term "meat quality trait" preferably means any of a group of traits that are related to the quality of the eating (or palatability) of the pig. Examples of such features include, but not limited to, muscle pH, loss of clearance (or ability to hold water), color of muscle, firmness, marbling scores, percentage of intramuscular fat, and softness. As used herein, the term "polymorphism" refers to the variation that exists in the DNA sequence for a specific marker or gene. That is, in order for a polymorphism to exist there must be more than one allele for a gene or marker. As used herein, the term "preconditioned conjugate gradient" preferably refers to a method for the symmetric positive definitive linear system. The method continues through the generation of sectors of repetition vector sequences, and the investigation of the addresses used in the update of the repetitions and residuals. As used herein, the term "depuration" (eg, "loin purification") preferably refers to the liquid escaping from the meat while it is in a vacuum sealed plastic package for a period of time (eg, during the first 7 days or during day 28). As used here, a "qualitative trait" is one that has a small number of different categories of phenotypes and for which the genetic component is generally controlled by a small number of genes. As used herein, the term "quantitative trait" is used to denote a trait that is controlled through a large number of genes each small for a moderate effect. Observations on quantitative traits usually follow a normal distribution. As used herein, the term "quantitative trait site (QTL)" is used to describe a site that contains polymorphism that has an effect on the quantitative trait. As used herein, the term "random genetic effects" is preferably used to denote factors with levels that were not deliberately organized by the experimenter (those factors are called fixed effects), but were, rather, sampled from a population of possible samples . A typical random genetic effect in animal husbandry is the additive genetic effect. In addition, random genetic effects can be subdivided into at least two categories. "Continuous random genetic effects" that are "quantitative" effects that are governed by a plurality of genes, each of which contributes additively to the quality or trait. "Discontinuous random genetic effects" are categorical or qualitative and may be dependent on a single site or few genetic sites. As used herein, the term "reproduction trait" refers to any of a group of traits that relate to the reproduction of the animal (e.g., reproduction of the pig and productivity of the sow). Examples in pigs include, but are not limited to, the number of piglets born per liter, the weight of the piglet at birth, the survival rate of the piglet, weaned piglets per liter, the weaning weight per liter, the age at which puberty, the rate of calving piglets, the days of estrus and the quality of the semen. As used herein the term "selection index" preferably refers to a sum of weighted EBVs for different economic features. The selection index for each animal is a relative value and can be expressed in biological or economic units. The animals are classified and selected based on the selection index. The values for the selection index are determined empirically and / or subjectively through the analysis of the market values for a given trait. For example, suppose that a trait for "efficient growth" is determined to have a potential future potential in the pig market and that two traits, body weight in 196 days (bw) and percentage without fat (Ip) are used as metrics for efficient growth. Also assume that through market analysis it is determined that each additional pound of bw of 196 days is worth $ 0.40 and each percentage point without additional fat is worth $ 2.00. In this model, the weights selected for bw and Ip are, respectively, $ 0.40 and $ 2.00. The selection index (I) is calculated according to the following equation: I = (0.4) (EBVbw) + (2.0) (EVB, P). Once EBV is calculated, the selection index can be used as part of a herd management program or system to identify the specific animals that are most likely to produce progeny that have the desired trait characteristics. It is observed that in order to be useful in the selection index the EBVs component must have been simultaneously calculated, on the contrary it will offer a different and not comparable scale.

PREFERRED MODALITIES OF THE INVENTION Several embodiments of the invention described herein provide the best unbiased assisted marker prediction (MA-BLUP) as part of the methods and / or systems that provide a fully integrated genetic evaluation system. The MA-BLUP methods and systems described here combine the methodology of the best traditional linear unbiased prediction (BLUP) with the theory of current marker assisted selection (MAS) in a robust individual computer executable algorithm useful for producing estimated breeding values (EBV) for each animal in a population. The theory and calculated algorithms described provide unexpectedly useful and effective extensions and modifications of previously known techniques. Several embodiments of the present invention provide algorithms of the best marker-assisted linear unbiased prediction implemented with MA-BLUP in a form that is functional and practical for use by breeding companies and / or large agribusinesses. The MA-BLUP methodology described here provides methods and / or systems that can be used to simultaneously analyze catches of pedigree data, production performance data, and genetic marker data of a population and produce EBVs for each animal in the population as an exit Among the unique characteristics of MA-BLUP as described herein is the ability to use molecular genetic information acquired from any method or form of genetic analysis including genotyping candidate genes (ie, genes of which certain variants are known or believed to be provide other economic advantages when present). Other methods of genetic analysis are well known to those skilled in the art and include, but are not limited to, marker genotyping (which may be based on RFLPs = restriction fragment length polymorphisms; simple sequence repeat (SSR, aka "microsatellite" markers), amplified fragments of the polymerase chain reaction (PCR), especially multiplex PCR (the simultaneous amplification of several sequences in a single reaction)) and single nucleotide polymorphism (SNP, which analyzes the differences in individual nucleotides in, for example, near a gene of interest). A particularly preferred aspect of the present invention is that it allows simultaneous analysis of three or more markers under multi-trait statistical models. In this way, the present invention provides methods and systems that allow those skilled in the art to evaluate a population of animals with respect to pedigree information and a preselected list of one or more quantitative traits, one or more QTL for each trait quantitative and three or more molecular genetic markers for each QTL. In addition, the methods and systems provided allow animals in the population to be categorized according to their EBV for a given trait or group of traits. Once the animals are classified, the classification information can be used as part of a breeding management system to achieve the desired breeding objectives. For example, it can be used to increase the average genetic significance of the population for the selected trait (s) and / or can be used to relatively quickly produce animals that have a genetic predisposition for highly expressed favorable of a pre-selected trait. Another powerful aspect of the present invention that will be appreciated by those skilled in the art is that the MA-BLUP invention can be modified to provide analysis of any type of population through the use of a variety of "statistical models". The various statistical models can be provided as input data in any of the embodiments of the present invention. The specifically statistical models are used to individually customize the general MA-BLUP methodology to adapt it to the specific data characteristics of the defined population. Thus, the present invention provides the general purpose MA-BLUP analysis that is independent of the statistical models that any particular user may wish to employ. For example, for the breeding of molecular pigs a major statistical problem is the estimated determination of breeding values for each animal in a population using data that includes pedigree information, metrics of farm animal traits (such as gain in average daily weight, litter size, average weight at weaning, and etc.), and molecular genetic data. A statistical model for this problem would be:? = Xb + Z1u + Z2v + e where? is a vector of phenotypic data, b is a vector of fixed effects, u is a vector of polygenic effects and v is a vector of QTL effects (site of quantitative trait). Variant-co-variant matrices are Gu for u and Gv for v. In addition, as will be apparent to those skilled in the art, statistical models for use with the present invention will also require parameters such as heritability of the selected traits and genetic co-relationships among the selected traits. Also, the distance between the markers and the speed at recombination between two markers are also important parameters for MA-BLUP. Another aspect of the various embodiments of the present invention is that the methods and systems described allow for effective "management of missing terms". This does not mean that all data must be provided for each animal in a population. For example, data may be provided for pedigree data for some animals but not for others.

Similarly, phenotypic or genotypic data (marker) may be missing for some individual animals but not for others. Thus, a powerful aspect of the present invention is that it allows the simultaneous analysis of several databases, including pedigree, phenotypic, and genotypic data that may have "missing" terms for any given animal. In this way, through the use of different statistical models the various embodiments of the present invention are specifically customized for methods, systems and, etc., to determine EBV for a wide variety of organisms including, but not limited to, animals. of farm, such as pigs, cattle, sheep, goats, poultry. Furthermore, it is well within the skill of one skilled in the art provided with the present invention to design a statistical model for use in any desired population, plant or animal. In preferred aspects of these modalities the population is formed of pigs, cattle, or sheep. In a particularly preferred aspect of this invention the population is a population of pigs. To assist in the speed and efficiency of the MA-BLUP analysis, several modalities of the invention use a gradient algorithm of pre-conditioned conjugate (PCCG) with diagonal blocking variable by size as a pre-conditioner. When the QTL effects are included in the linear mixed model, we first find that it is more effective to take the block diagonal n for n for the polygenic portion and 2n for 2n for the block diagonal for the QTL portion in the linear equation systems as a pre-conditioner, where n is the number of features in the analysis. The pre-conditioning strategy is referred to as a "diagonal pre-conditioner block of variable size" algorithm. The comparison with the preconditioning algorithm that was previously used in common computer packages the variable diagonal block preconditioning algorithm is 150% more effective in terms of computing time. This dramatically reduces computing time. Pre-conditioning is a technique commonly used in linear algebra. For example, suppose that you want to solve the following linear equation: Ax = b. A pre-conditioner is a matrix, "M". The pre-conditioning procedure involves multiplying both the side of the linear equation by M, which is MAx = Mb. It is observed that this pre-conditioning procedure has two characteristics: it does not change the solution and makes the resolution procedure faster and the resolution more exact (see Shewchuk, 1994). Equation 1, below, provides the pseudo code of an algorithm for solving the Ca = r problem using the preconditioned conjugate gradient method, as provided in Strandén, I. and M. Lidauer, 1999, which is incorporated herein by reference.

Equation 1 a (0) = initial estimate; r0 (0) = r-Ca (0) d (0) ^ M-1 r0 (0); / o = ro (0) d (0), for? = 1,2, ... q (k > < = Cd < k "1 >; ak = / k-1 / d (k) q < k > a'k»: a (k-1) + akd (k-1 »if K is divisible by 100 r0 (k) < rk (k) otherwise r0 (k) = r0 (k-1) - ak q (k) SW ^ M -1 roW / k «= r0 (k) s (k); ßk < = / k // k-1 d (k) = s (k) + ßk d (k) if there is no convergence continue with the repetition end. The "M" employed by the various aspects of the present invention is a diagonal block matrix. For the present example, assume that there are features t. "M" consists of three parts: y = Xr -Zl? u + Z2v + e (a) the blocks t by t extracted from the diagonals of the following (a block is a subgroup from the left hand side of the mixed model equation). X 'R-1X (b) the blocks t by t extracted from the diagonals of the following (c) blocks 2t by 2t extracted from the diagonals of the following Z2' R'1 Z2 + G? Although the previous BLUP programs implemented the data repetition algorithms (lOD), these previous programs were only 50% effective according to that provided by the present invention. This is because the 'pre-calculated and stored' algorithm implemented in the present invention. The steps that were time consuming, but independent of the repetition steps on data (such as individual calculation of the contributing coefficients when calculating the inverse of the variant-co-variant matrices for QTL) are pre-calculated and stored for later use in each repetition. An optimized order of matrix-vector multiplication is implemented in lOD. In addition, as described here applicants have created methods and systems to apply and integrate variable blocking algorithms and PCCG algorithms with repetition on the data to provide a surprisingly useful and powerful analysis of molecular genetics, character traits, pedigree information, animal that provides those involved in the administration of the population of animals with effective means to verify and evaluate EBV for individual animals. These evaluations can then be used as part of a herd management system. Additionally, the various modalities of the present invention employ the data repetition methodology, which greatly reduces computer memory requirements. The animals may be selected for use in accordance with the present invention through any suitable means. For example, using computer programs or other means to register the lineage / pedigree and select the most suitable equipment. The use of computer programs can also be improved with the capture of biometric data including the use of molecular genetic analysis. The methods and systems of the various embodiments of the present invention use computer algorithms to solve mixed model equations (MME) that are taken into account and provide the output to guide breeding based on both fixed and random genetic effects (! including both continuous random effects and additive genetic effects, and discontinuous or categorical random effects). The various embodiments of the present invention provide methods to improve an estimated breeding value of an animal population or to identify breeding pairs in order to rapidly maximize the manifestation of a desirable trait. That is, the methods and systems of the present invention can be used to identify those potential parent animals that, when raised with one another, will most likely manifest a maximal improvement of the selected trait in their progeny. In accordance with the various aspects of this embodiment of the present invention, the method comprises: (1) selecting one or more trait (s) for the population where improvement is desired. (2) Provide the animal population with a database that contains data on one or more sites of quantitative traits. (3) Provide the data base for individual animals in the population where the database comprises data for one, two, three, or more molecular genetic markers for each QTL, for each trait for which an improvement is desired. (4) Provide a database that includes the pedigree data for the animals in the population. (4) Optionally provide data regarding the physical effects for the animals in the population. (5) (6) Provide and use a computer program capable of carrying out the best unbiased linear prediction assisted by marker to concurrently analyze the data of the provided databases and to calculate and provide as an output of those calculations, a estimated breeding value (EBV) for each of the animals for the selected traits, and a classification of the animals with respect to their individual estimated breeding values. A particular aspect of this embodiment of the invention provides for the use of EBVs to prepare a breeding plan for the animal population that provides an optimum improvement in the average of the genetic importance of the population or to maximize the genetic importance of a specific progeny. . In any aspect of the invention the number of traits selected and the number of quantitative trait sites (QTL) for each trait may be one or more. In a preferred aspect of the invention the number of QTLs selected for each feature can be 1, 2,3,4,5,6,7,8,9,10,15,20, or 30, or more. In addition, in any aspect of the invention, the number of molecular genetic markers for each QTL can be 1, 2,3,4,5,6,7,8,9,10,11, 12,13,14,15, 16,17,18,19,20, 25 or 30, or more. In preferred aspects of any modality of the invention the number of molecular genetic markers is 2 (two) or more. In even more preferred aspects of this embodiment, the number of molecular genetic markers is three or more. In preferred aspects of this embodiment of the invention, markers linked to QTL can form a haplotype. In this sense, a marker haplotype is a particular group of marker alleles of two or more neighboring markers that tend to be co-inherited. To be co-inherited, the markers that make up the haplotype can be located relatively close to each other (for example, all markers could be located within a range of 5 cM). In an even more preferred aspect of this embodiment, to increase the likelihood of coherence, the haplotype-forming markers are located within a smaller interval of cM in width. As an example, if 3 SNP markers were located close enough to be co-inherited, and if these markers have the following possible alleles, Then, the possible haplotypes could be as follows: ACÁ, ACC, AGA, AGC, TCA, TCC, TGA, TGC. These individual haplotypes can be inherited by several generations with little opportunity for recombination and, therefore, can vary significantly in terms of their lineage to possible QTL alleles. As the number of alleles increases by marker, or the number of markers per haplotype, the number of possible haplotypes also increases, but in an exponential way. Therefore, the ability of the BLUP methods and systems described here to include several markers per QTL increases the informativity of the marker haplotypes linked to QTL, thus greatly increasing the probability of finding linked markers, as well as the probability of precisely track QTL alleles marked in successive generations. In addition, the ability to use marker haplotypes increases the flexibility and robustness of the MA-BLUP program described here. In any aspect of this embodiment of the invention the type of molecular genetic markers can be selected from, but not limited to, the group consisting of: RFLPs (restriction fragment length polymorphism), simple sequence repeat (SSR, aka "microsatellite" markers), fragments amplified with polymerase chain reaction (PCR), especially multiplex PCR (the simultaneous amplification of several sequences in a single reaction) and individual nucleotide polymorphisms (SNPs), which detect differences of individual nucleotides in, for example, a gene of interest). Marker information can also include data on point mutations, deletions or translocations or other gene isoforms. According to a particularly preferred aspect of this embodiment of the invention, the marker is selected from the group consisting of SNPs of the porcine PRKAG3 gene, variants in the porcine leptin receptor gene (pLEPR), and the melanocortin-4 receptor (MC4R). The melanocortin-4 receptor (MC4R) is described in three references each of which is incorporated herein by reference. These references include: (1) Kim and others Mammalian Genome (2002) 11 (2): 131-5, which indicates that a missense variant of the porcine melanocortin-4 receptor (MC4R) gene is associated with traits of fat, growth, and food intake. (2) WO 00/06777 (Rothschiid et al., Indicates that MC4R is a marker for growth and input of food and fat content). A polymorphism (a missense mutation Asp298 caused by a substitution of individual nucleotide G678A) in the MC4R gene was identified and found to be associated with the growth rate, the food intake, and the fat content in the body . A detection method based on RFLP is described and used for genotyping. Additionally, the detection method based on A TAQMAN® is contemplated through the invention to detect the individual nucleotide polymorphism. (3) WO 01/075161 (Rothschild et al., Describes MC4R as a marker for meat quality traits). The polymorphism (G678A) in the MC4R gene is described as being associated with several meat quality traits including pH, drip loss, marbling and color in the pig. A detection method based on RFLP for genotyping is described here. In any aspect of this embodiment of the invention the computer program may be configured to provide an evaluation of the "nformativity" and / or approximation "of each molecular genetic marker with respect to the trait for which it serves as a marker. methods and systems of the present invention can be configured to determine which marker or markers are the most "informative" and which are the most approximate for the site of the quantitative trait for which they serve as a marker.The porcine leptin receptor gene (pLEPR) has been located on chromosome 6, in approximately 122 cenfiMorgans (cM). In addition, a number of DNA sequences (genomic and cDNA) for the porcine LEPR gene are available from the Genbank public DNA database including: accession numbers: AF092422, AF167719, AF184173, AF184172, AH009271, AJ223163, AJ223162, U72070, AF036908, and U67739 each of which is incorporated herein by reference. It has been shown that a useful allelic polymorphism comprises a "C / T" variation in the fourth exon of the leptin receptor gene. This variation results in the pLEPR protein produced from these variants having either methionine or a threonine as a 69 amino acid of the pLEPR prepro protein (see, Figure 7). The C / T polymorphism results in either a variant cytosine ("C") or thymine ("T") in the nucleotide corresponding to position 609 of the Genbank registration number AF184172 in the fourth exon of the pLEPR gene. This polymorphism produces a pLEPR protein that has either methionine (if the nucleotide is "T") or a threonine (if the nucleotide is "C") in the amino acid number 69 of the pLEPR prepro protein. The "T" variant (containing thymine, which encodes methionine) is believed to be the most common. Common abbreviated designator, the polymorphism will be referred to as the "T69M" polymorphism. An analysis of 2626 pigs from an individual commercial line, shown in the presence of the "C" allele that had a statistically significant correlation with a positive effect on: ADG early (average daily gain from day 0 to day 90 of life); Late ADG (average daily gain from day 90 to day 165 of life), the pH of the loin muscle, the color of the loin muscle, and the loss of drip. There was a small negative effect on the "C" allele in the back fat, that is, the fat in the back increased slightly.

In addition, ninety-seven (97) SNP markers that represented 38 sites on pig 6 chromosome (SSC6) were typed on a panel of 1, 444 pigs from the pure line of the commercial line. The site selected for the SNP discovery was separated through approximately an 80 cM region in SSC6, which included the LEPR site and the SNP produced the T69M mutation. Linkage disequilibrium analysis was used to identify both haplotypes SNPs and SNPs (for up to three adjacent sites) that were significantly associated with phenotypes related to growth (ie, thickness of back fat, flaccidity, weight off the test, and weight gain). The 97 SNPs and possible combinations of two or three adjacent SNP haplotypes were evaluated for association with all phenotypes. Only four SNPs (plus several haplotypes containing these SNPs) were found to be significantly associated with the thickness of the back fat, they were corrected for any age or weight. One of these SNPs included T69M and the other three were represented within 3 cM of T69M as estimated by linkage analysis. Accordingly, the present invention can be employed using a marker for the T69M mutant pLEPR, or any marker in the disequilibrium linkage with said marker. In any embodiment of the present invention the MA-BLUP program used may be integrated with a "programming feature" that allows the user to manipulate the algorithms of the program using a programming language that is similar to common English. For example, if the program that implements MA-BLUP is written in the C ++ computer programming language, the programming feature allows the user to use the MA-BLUP program without knowing C ++. The presently described MA-BLUP provides methods and systems that allow those skilled in the art to analyze a collection of one, two, three or more markers for a given quantitative trait site and determine the informativity of various markers. As noted in the definitions section, the "informativity" of a given marker provides an indication of how likely it is that an animal inherits the marker and will also express the desirable trait associated with that marker. Prior to the creation of MA-BLUP as used in the present invention, the best that can be said was the presence of the indicated marker at a 50:50 chance that the desirable trait might be present. By providing means for quantifying the informativity of a given marker or group of markers, the presently described methods and systems provide a better forecasting tool, the present invention provides methods and systems for determining which of a group of markers is the best predictor for a particular feature (ie, the most informative) and provide an indication of the proximity or proximity of the marker to the quantitative trait site associated with a given trait. Various embodiments of the present invention provide methods and systems for increasing an average genetic significance of animal populations for one or more pre-selected traits. The various embodiments of the invention also provide systems for rapidly improving a given trait in a progeny by providing means to select those animals within the population that will most likely effectively pass the germplasm to express the trait to their progeny. The systems according to this aspect of the invention comprise the following components. (1) A suitable computer to allow the capture of database and / or the execution of a program to calculate the EBVs of the animals using the methods described here and providing access to the user and an interface with the computer. (3) A database or databases accessible by computer that provide individual data for each animal in the population for each of one, two, three or more molecular genetic markers for a particular quantitative trait. (4) A computer-accessible database that provides individual pedigree data for each animal in the population. (5) Optionally a computer-accessible database that provides individual data for each animal in the population for at least one feature of interest. (6) A computer executable program capable of using MA-BLUP to simultaneously evaluate data in all databases and to classify animals in the population according to their respective estimated breeding value. (7) A user interface preferably including a data entry system, said user interface coupled to said computer and configured to allow the user to instruct the computer to access the available databases and use the computer program MA- BLUP to generate an output of the EBV classification of the animals and / or their estimated individual breeding values. In preferred aspects of this modality of the invention, the population of animals is selected from a herd of pigs, herd of bovines, and a herd of sheep, before the systems to evaluate any type of plant or animal population is contemplated as falling within the present invention. In a particularly preferred embodiment the system is designed to evaluate the estimated breeding values for the pig herd. Those skilled in the art will appreciate that the methods and systems of the present invention can be used to evaluate any type of molecular genetic marker. Accordingly, any specific marker disclosed herein is intended to be illustrative only and not to limit the scope of the invention in any way. Despite this fact, in particularly preferred embodiments of the invention the markers are selected from those that measure the variation in the porcine PRKAG3 gene, the porcine leptin receptor gene, and the MC4R gene. In all embodiments of the invention methods and systems can be used to evaluate a BV of the animal population for a defined set of features. In addition, these methods and systems can be used to identify those individual animals or groups of animals that optionally provide the germplasm necessary to improve the frequency and / or quality of the desired trait. Meaning that the breeding pairs can be selected to optimize the expression of the selected trait in the progeny of the animals. Other embodiments of the present invention also provide for the analysis and quantification of the predicted value relative to the site markers of the quantitative trait. The invention provides methods and systems that compute the informativeness and / or approximation of a molecular genetic marker to the site for the trait which serves as a marker. In addition, with respect to the quantitative trait markers the methods and systems of the present invention also provide an indication of the informativeness of the marker. Several embodiments of the present invention also provide for the use of markers described supra. That is, the present invention provides as one of its aspects, means for using markers to identify those animals suitable for use in accordance with the invention. This procedure is called MAS (marker-assisted selection). The invention also contemplates the use of MAA (marker-assisted allocation). Through the use of MAA, the selected animals are assigned for use to more effectively and efficiently bring the desired genetic improvements in the progeny of animals. In certain embodiments of the present invention, the information / data obtained from the analyzes of various biometric measurements as well as other types of information (eg, pedigree) can be weighted in a "selection index" in order to provide an evaluation of a value of the animal as a father, that is, its estimated breeding value. Phenotypic measurements are affected (biased) through the herd and the year or season in which the performance of the animal is measured. In order to correct this bias, a procedure called BLUP (Better linear unbiased prediction of breeding value) was developed (see, Animal Breeding, p.84). As noted above, there are currently several computer programs available from the software authors that can be used to calculate the values that can be used to calculate the BLUP values. Inbreeding is defined as the probability that two genes (ie, alleles) in one site are identical by descent (Malecot, 1948).

The level of inbreeding (Fx) (ie, inbreeding coefficient) can be calculated from the pedigree records that go back to the founding animals of a given population as follows: (where, a? S? D is the genetic relationship additive between Xs and Xd, if X is the progeny of Xs and Xd) The increased homozygosity due to inbreeding is generally perceived as having detrimental side effects such as depression of inbreeding (ie, a decrease in production yield , reproduction, and fitness traits) and decreased genetic variation leading to reduced rates of genetic gain over time.

The rate of inbreeding,? F, is defined as the increase in the inbreeding coefficient in one generation (Falcaner and Mackay, 1996), and can be approximated by:? F = 1/8 Nm + 1/8 Nf In where Nm and Nt are the number of males and females, respectively, that contribute to the next generation. As is evident in this approach, while fewer animals are selected as parents, the rate of inbreeding tends to increase. Unfortunately, the pressure of increased selection takes the form of selecting a smaller proportion of parents for the next generation. Therefore, pig raising companies usually try to balance extra genetic gain from the selection of fewer parents against the resulting increase in inbreeding speed. Typically in pig populations, many females are selected to produce enough progenies for the next generation; therefore, inbreeding caused by female parents is not usually a concern. However, in order to limit the degree of inbreeding and maintain genetic variation in the herd it is common practice to select more males that are strictly necessary for breeding purposes. This practice limits both the degree of genetic advance GN and the speed at which changes can be made in the genetic frequency and the direction of the trait. When several progenies should be selected as a parent, it is difficult to find a progeny group that has high breeding values with a particular genetic profile (for example, the profile of the specific genetic marker).

Limitations due to multi-trait selection indices: Typically, selection in a population is practiced through the use of a multi-trait selection index. In this method, estimated breeding values are calculated for each economic trait for each animal based on pedigree and phenotypic information. The estimated breeding values are then weighted according to the relative economic value of each trait as well as the direction provided for selection for the population and are incorporated into an individual, multi-trait selection index. These multi-trait indices incorporate several sources of information for each animal (for example, phenotypic records of ancestors, progeny and the animal itself). Selection rates determine progress, the long-term genetic advance for the population and must be carefully constructed to balance the needs of both present and future markets. Therefore, if temporary changes occur in the market, the parent company can not fully justify and change the selection index to reflect those changes; especially, if the future market conditions will probably not coincide with the current, temporary conditions.

Two-stage selection Typically, the selection takes place in quantitative traits based on the BLUP breeding values and is classified into a multi-trait selection index. However, there is an increasing number of economic feature sites (ETLs) that have been discovered and reported that are associated with traits that are not normally considered in the multi-feature selection index and that have a measurable economic value. (for example, health traits or quality of the meat). A simple method to use these genes is through the selection of two stages. In the first stage, the animals could be genotyped for one or more ETL after being pre-selected for the most favorable form (allele) of ETL. Then, in the second stage, an additional selection is made on the rest of the animals according to the traditional multi-trait selection index. This method has the benefit of being relatively easy to apply and can reduce the number of animals for which phenotyping is necessary (for example, gain on the test, ultrasound measurements of the fat on the back and the eye area of the eye). loin, etc.). Alternatively, the first step may comprise standard phenotyping procedures, and classifications in accordance with multi-trait MA-BLUP EBVs. This is then followed by a second stage in which the animals are differentiated according to their genotypes in one or more ETL. This second option does not present any savings in phenotyping, but could provide savings in genotyping if some animals are classified too low to be considered for selection, and therefore, genotyping costs are not justified. In addition, some genotypes may have more value for certain clients than for others, and therefore, marker-assisted allocation (MAA) can be used to assign specific animals to clients that desire a particular genotype. MAA can therefore be justified through the charging of a premium to customers who receive the specified genotype.

Individual stage selection (multi-scratch index) By simultaneously incorporating all the information available at the time of selection, in the form of a single-stage multi-feature selection index, it is the most efficient form of selection. In addition, this method results in a longer long-term advance towards the stipulated breeding objective. Other selection strategies such as the selection of two stages (previous) the joint selection (that is, alternative selection on different traits over multiple generations) or the use of independent selection levels (ie, eliminating animals that do not achieve a threshold of minimum selection) have shown that they are less efficient than index selection (Van Vleck, et al., 1987). However, these other methods are sometimes used for reasons related to ease of use cost or implementation speed. The index selection usually takes the form of a linear equation as follows: H¡ =? -? Aij +? 2 A2¡ + ... +? NAN¡ where H¡ is the selection index value for the animal i ,? -i,? 2 and? N are the net economic values per unit of trait 1 through N, Ap, A2¡ and AN¡ and the genetic value additive for the animal / for traits 1 through N. Genetic values Additives for each trait can be calculated to include the ETL information through MA-BLUP (described above). Additional information is readily available with respect to index selection (Van Vleck et al., 1987; Van Vleck, 1983). One of the most difficult aspects of incorporating information ETL in the multi-feature index selection is to determine how to properly weigh the new information in relation to the phenotypic information of the traditional trait. Since ETL information is usually conditional on marker genotype information, this information may be difficult to include, because markers are usually not located directly in ETL, but rather there is some distance between them. Recombination (chromosomal crosses) can divide the bond (association resistance) between the marker and the ETL, and tends to incur in proportion to the distance and the marker and the current ETL. This degree of recombination needs to be taken into account as well as situations in which genotypes are not available in all animals. This procedure has become very reliable with the arrival of the MA-BLUP methodology. (see above), where the ETL information is combined with the additive genetic breeding value for the trait for the animal. In the MA-BLUP scenario, marker information can be simultaneously included with phenotypic and pedigree information to predict breeding values. If the trait affected by ETL is already included in the multi-trait selection index, then the classification and selection may continue more or less as previously described. However, if the ETL affects a new feature that is not currently in the parenting goal, additional work must be done. First, evaluate the economic value of the new trait and, secondly, estimate the necessary genetic parameters that surround the new trait (ie, heritability, genetic variant and co-variant with the other traits in the selection objective). The information regarding the estimated genetic parameters and the applications of the BLUP models used in the breeding of animals is known to those skilled in the art (see, for example, Henderson, 1984).

PRKAG3 The PRKAG3 gene encodes the gamma sub-unit of porcine AMPK (protein kinase activated by adenosine monophosphate), whose enzyme has been shown to play a key role in the regulation of energy metabolism in eukaryotic cells (Milan et al. 2000) . Animals that have certain variants in the PRKAG3 gene have shown that they have more desirable characteristics with respect to the pH of loin and ham, to have a reduced 7 day purge from the loin muscle, to have a reduced drip loss, and other quality traits of the meat. According to several embodiments of the present invention MA-BLUP can be used to classify the EBV of the animals in a population of pigs based, inter alia, on the animal complement of several PRKAG3 SNPs. That is, based on the haplotype of the animal for the PRKAG3 gene. In accordance with the various aspects of this embodiment of the invention, the EBV classifications of the herd population are then used as part of the herd management / breeding program useful to improve the average genetic significance of the quality traits of the herd. meat in general and specifically with respect to meat quality traits influenced by the PRKAG3 haplotype of the animal. Several embodiments of the invention provide methods, equipment, and compositions that are mapped for use of porcine PRKAG3 gene SNPs. Aspects of this embodiment of the invention are useful for improving one or more meat quality traits. The improved meat quality traits include all those commonly measured by those skilled in the art. In preferred aspects of this embodiment of the invention the meat quality traits are selected from the group consisting of increased backbone, increased ham pH, reduced 7 day purge and reduced trickle loss. Certain aspects of this embodiment of the invention provide methods to improve the quality traits of the meat of animals in a herd and / or to classify the plurality of animals in a herd to identify the nature of the PRKAG3 haplotypes present in the classified animals. Then these pigs identified as having one or more desired alleles are used as part of a breeding plan to produce progeny that has an increased frequency of the desired allele and / or trait. In a preferred aspect of these embodiments, the SNPs are selected from one or more known SNPs in the porcine PRKAG3 gene. In a more preferred embodiment of the invention the SNPs are selected from the group consisting of: an A / G at position 51, A / G at position 462, A / G at position 1011, C / T at position 1053 , C / T at position 2475, A / G at position 2607, A / G at position 2906, A / G at position 2994, and C / T at position 4506 (note that the numbering provided above is from according to the sequence of SEQ ID NO: 1). It is noted that the selection procedure may include the use of the MA-BLUP program described herein. Any suitable method for classifying animals for their status with respect to the recently described PRKAG3 polymorphisms is considered as part of the present invention. Such methods include, but are not limited to: DNA sequencing, restriction fragment length polymorphism (RFLP) analysis, heteroduplex analysis, conformational polymorphism analysis of individual chain structure (SSCP), denatured gradient gel electrophoresis ( DGGE), real-time PCR analysis (TAQMAN®), temperature gradient gel electrophoresis (TGGE), primer extension, allele-specific hybridization, and INVADER® genetic analysis assays.

EXAMPLES The following examples are included to demonstrate the preferred embodiments of the invention. It should be appreciated by those skilled in the art that the techniques described in the examples that follow represent techniques discovered by the inventor to function well in the practice of the invention, and are thus considered to be the preferred modes for their practice. However, those skilled in the art, in light of the present disclosure, should appreciate that many changes can be made in the specific embodiments described and still obtain a similar or equal result without departing from the invention.

EXAMPLE 1: Marker mc4r used in the line of commercial pigs Of approximately 600 young animals from a performance test station, the 10 main males were selected for incorporation into the breeding herd to produce the next generation of animals.

Phenotypic data animal sex litter cgp old wda Flacura p 0000001016391 M 20047 90006 160 109 0000001030745 M 20048 90006 164 552 0000005010960 M 20049 90172 170 169 500 0000005010985 M 20050 90172 174 141 536 0000005010986 M 20050 90172 167 141 515 0000005010987 M 20050 90172 174 118 545 0000005011018 F 20050 90172 167 113 601 0000005011019 F 20050 90172 167 113 515 0000005011020 F 20050 90172 167 119 552 0000005011021 F 20050 90172 167 106 546 2220000007490 M 34789 90682 154 103 492 2220000007494 M 34789 90682 154 127 511 2220000007497 F 34789 90682 154 115 533 2220000007498 F 34789 90682 154 96 520 2220000007499 M 34790 90682 154 131 525 2220000007501 M 34790 90682 154 140 534 2220000007503 F 34790 90682 154 136 511 2220000007505 F 34790 90682 154 110 508 2220000006486 F 34796 90682 152 124 531 2220000006487 F 34796 90682 152 80 556 Genotypic data animal genotype 0009705450992 A / G 0009705451278 A / G 0009705451281 A / G 0009705451282 A / G 0009705451288 A / G 0009705456787 G / G 0009709501525 A / G 0009709501528 A / G 0009709501530 G / G 0009709501531 G / G 2220000006032 A / G 2220000006033 A / G 2220000006034 G / G 2220000006035 A / G 2220000006036 A / G 2220000006037 G / G 2220000006038 G / G 2220000006039 G / G 2220000006040 A / G 2220000006041 G / G Pedigree data Animal progeny female sex 0000009000347 0000009000345 0000009000245 0000009000351 0000009000352 0000009000346 M M M 0000009000367 0000009000361 0000009000366 0000009000350 0000009000348 0000009000363 0000009000361 0000009000362 0000009000349 M M M 0000009000365 0000009000269 0000009000364 0000009000358 0000009000347 0000009000344 0000009000221 0000009000276 0000009000357 M M M 0000009000360 0000009000227 0000009000359 0000009000334 0000009000269 0000009000333 M 2220000008593 1090000024220 1090000021806 2220000008594 1090000024220 1090000021806 F F F 2220000008595 1090000024220 1090000021806 2220000008596 1090000024220 1090000021806 2220000006876 1130000051724 1090000024984 F M M 2220000006877 1130000051724 1090000024984 2220000006878 1130000051724 1090000024984 M 2220000006879 1130000051724 1090000024984 2220000006880 1130000051724 1090000024984 F F F 2220000007516 1130000051724 1100000031328 Statistical Model There are two traits: weight per day of age (wda) and percentage of flacura (leanp). wda = age age * age sex cgp mc4r animal breeding leanp = age age * age sex cgp mc4r animal breeding Animal Classification EXAMPLE 2 Identification of new snps in the prkag3 gene and its use to improve ebv for meat quality traits in pig herds The porcine PRKAG3 gene was expressed exclusively in skeletal muscle and was involved in the regulation of glycogen synthesis. There is now convincing evidence in the technique that supports the hypothesis that mutations in this gene affect meat quality traits such as glycolytic potential (GP, is an indicator of the level of glycogen in a live animal that is calculated as the total of the total main compound susceptible to conversion to lactate, GP equal to 2 (glycogen + glucose + glucose-6-phosphate) + lactate), pH, drip loss, and purification. At least two different single nucleotide polymorphisms (SNPs) that alter the amino acid sequence of the mature protein have been found in exons for this gene. In addition, these polymorphisms have shown that they are associated with the meat quality traits listed above. For example, there are two separate international patent applications (WO 01/20003 A2 and WO 02/20850 A2) drawn for the use of these SNPs. We describe here nine (9) recently identified PRKAG3 SNPs that have been shown to be associated with meat quality traits. The porcine AMPK sequence (AMP-activated protein kinase) available with accession number of Genbank AF214521 (see Figure 4) was used to prepare the primers for use of the amplification fragments represented by most of the sequences known for this gene (see Table 1 for the primer sequences).

TABLE 1 Names of the primers and sequences used to amplify PRKAG3 for the discovered SNP.

The genomic DNA of twelve (12) unrelated animals from a commercial "A" line of pigs was used as the template for the amplification using eight pairs of primers, set forth in Table 1 as primers. After amplification, the resulting amplicons were sequenced and the sequence of the twelve animals was aligned, amplicon by amplicon, and evaluated to identify potential sequence polymorphisms. Twenty-four (24) SNPs were identified including several of the SNPs identified in the patent applications (WO 01/20003 A2 and WO 02/20850 A2). TAQMAN® SNP assays were designed and validated for eleven SNPs, including nine SNPs that were previously unknown, (see Table 2).

TABLE 2 SNPS PRKAG3 whose TAQMAN® assays were successfully validated.

These SNPs were then genotyped on a panel of 2,693 animals from two different commercial lines, "A" and "B", representing 118 half-sib families with meat quality phenotypes. The SNP haplotypes were determined for as many animals as possible and the association analysis was carried out to determine which haplotypes were the most predictive / informative for the various quality traits of the meat. Although there are theoretically 211 different possible haplotype groups with 11 different SNPs, almost 95% of the animals for which the haplotypes could be completely determined had one of only three different types of haplotypes (see Table 3). A particular haplotype (group 2 Hap.) Was significantly (p <0.001) associated with an increased pH in both the loin and the ham. In addition, this group 2 Hap. It was also associated with a reduced 7-day debulking of the loin muscle (see tables 4 and 5).

TABLE 3. Main SNP haplotypes for the eleven PRKAG3 SNPs genotyped in the population panel of the commercial pig line A '.

TABLE 4. Estimated average allele effect for the haplotype of groups 1, 2 & 3.

TABLE 5. Impact of haplotype fixation As can be seen from Table 3, which shows the three major haplotype groups, all SNPs with the exception of c1845t (SNP trial 148004) were in an almost complete link imbalance with each other. In this way, a genotype of any of the 10 SNPs (apart from c1845t) were genotyped in PRKAG3 as predictors, with a high degree of confidence, of the genotype in any of the other nine SNPs. Figures 5 and 6 show genotype and breeding values, respectively for c1845t SNP (SNP assay # 148004) and a2906g SNP (SNP assay # 148009), which are representative of ten SNPs with an almost complete link imbalance. The favorable allele of 148004 for an increased pH and decreased clearance of 7 days is the "A" allele, while the favorable allele for these features for 148009 is the "G" allele. As demonstrated by these figures (and also by Table 6) 148004 counts for a higher degree of variation in meat pH was 148009 (ie, it is either a casual mutation or a larger linkage disequilibrium with casual mutation ). However, the selection for the G allele of 148009 (or of the favorable alleles of nine other markers was found to be a linkage disequilibrium with 148009), can also be used to select animals in the commercial line A for quality traits of meat improved pH and depuration of 7 days.

TABLE 6. Gene effects and breeding values for SNPs 148004 (004) and 148009 All methods described and claimed herein may be made and executed without undue experimentation in light of the present invention. Since the compositions and methods of this invention have been described in terms of preferred embodiments, it would be apparent to those skilled in the art that variations can be applied to the methods and steps or step sequences of the methods described herein without departing from the concept of the invention. More specifically, it will be apparent that certain agents that are both chemically and physiologically related can be substituted by agents described herein while the same or similar results can be achieved. All such similar substitutes and modifications apparent to those skilled in the art are deemed to be within the scope and concept of the invention as defined in the appended claims.

EXAMPLE 3: PRKAG3 marker used in a commercial line of pigs A ' The analysis was done on 60 boars that left the performance testing station in March 2003. The top ten of these were selected for introduction into the breeding herd to produce the next generation. Two SNP markers were used in MA-BLUP for the following calculations.

Phenotypic data animal female sex lineag breeding cgp cgp3 old WdB line? 1 P 0000000628060 0000000103005 F 16 21597 90442 0 152 -r ^ jO 501, 0000000499339 aa00GQO4S245l F 15 21600 90442 ü 151 154 502. 0000000499340 0000000452451 F 15 21600 90442 0 151 132 511, 0000000499341 0000000452386 F 15 21601 90442 0 151 149 463, OQQQ000499342 0000000452336 F 15 21601 90442 0 3.51 123 454 ß 0000000499343 0000000452270 F 15 21602 90442 0 151 137 510 # 0000000499314 0000000452747 F 15 21603 90442 0 150 147 472 00-00000499315 0000000452747 F 15 21603 90 42 0 150 333 487. 0000000499316 0000000452010 F 15 21604 90442 0 150 145 456.

OOOOD0049S317 0000000452010 F 15 21604 90442 0 150 143 502 - 1070000010847 113000005-6726 F 16 32609 90422 699 172 140 501 610 1O7OD0D010875 1X30000054850 F 16 32810 90422 699 172 145 528 634 1070000010877 1130000054850 F 16 32810 90422 699 171 148, 602 1070000010899 1130000056330 P 16 32811 90422 699 171 143 499 604 1070000010901 1130000055380 F 16 32811 90422 0 171 137 485" 1070000010903 1130000056380 F 16 32811 90422 699 171 143 496 607 2220000002623 1090000025314 F 15 32833 90505 0 17 S 112 543, 2220000002624 1090000025314 F 15 32813 90505 0 178 116 552. 2220000002625 1090000025314 F 25 32813 90505 0 178 83, t 2220000002626 1090000025314 F 15 32813 90505 0 178 112 544 _ Genotypic data Animal m004 m009 0001995120096 G / GG / G 0001996264361 G / GA / G 0001996229682 G / GG / G 0001996237608 G / GA / G 0009645400235 A / GG / G 0009645408986 G / GA / G 0009652443262 G / GG / G 0009652443205 A / GG / G 0009652450481 G / GA / G 0009652424155 G / GA / G 2220000005567 A / GA / G 2220000005568 A / GG / G 2220000005569 A / GG / G 2220000005570 G / GA / G 2220000005572 G / GA / G 2220000005572 G / GA / G 2220000004935 G / GG / G 2220000004936 G / GG / G 2220000004937 A / GG / G 2220000004938 A / GG / G Pedigree data animal breeding female sex 0000000449871 0000000449568 0000000449554 M 0000000449875 0000000449568 0000000449554 F 0000000449876 0000000449568 0000000449554 F 0000000449877 0000000449568 0000000449554 F 0000000449878 0000000449565 0000000449562 M 0000000449877 0000000449565 0000000449562 F 0000000449881 0000000449565 0000000449562 F 0000000449872 0000000449564 0000000449563 M 0000000449879 0000000449564 0000000449563 F 0000000449882 0000000449564 0000000449563 F 2220000006808 1090000024991 1130000054009 F 2220000006809 1090000024991 1090000024710 M 2220000006810 1090000024991 1090000024710 M 2220000006811 1090000024991 1090000024710 M 2220000006812 1090000024991 1090000024710 M 2220000006813 1090000024991 1090000024710 M 2220000006814 1090000024991 1090000024710 F 2220000006815 1090000024991 1090000024710 F 2220000006816 1090000024991 1090000024710 F 2220000006817 1090000024991 1090000024710 F Statistical model wda = age sex line g cgp breeding animal flacura p = age sex line g cgp breeding animal pH = line g m004 cgp3 female and animal Classification of animals SSR markers used in the research line 79 boars that came from the performance test station in March 2003. The top 10 of these were selected in the breeding herd to produce the next generation. The markers 26 QTL and 55 SSR were used in MA-BLUP to select the 10 main boars.

Pedigree data female animal breeding sex 0000000449554 0 0 0000000449558 0 0 0000000449562 0 0 0000000449563 0 0 0000000449564 0 0 0000000449565 0 0 0000000449566 0 0 0000000449568 0 0 0000000449573 0 0 0000000449579 0 0 sex female animal breeding 1130000062981 1020000011792 1020000012539 1130000062981 1020000011792 1020000012539 1130000062981 1020000011792 1020000012539 1130000062981 1020000011792 102.0000012539 1130000062981 1020000011715 1020000011830 1130000062981 1020000011715 1020000011830 1130000062981 1020000011715 1020000011830 1130000062981 1020000011715 1020000011830 1130000062981 1020000011715 1020000011830 1130000062981 1020000011715 1020000011830 Statistical model bf = sex cg196 edad196 breeding mc4r_a mc4r_d bf_ql bf_q5 bf_q6 bf_q12 bf_q16 Animal lea = cg sex 196 age 196 breeding mc4r_a mc4r_d Iea_q2 Iea_q3 Iea_q7 Iea_q8 Iea_q12 Wt = sex cg196 edad196 breeding mc4r_a mc4r_d wt_ql wt_q2 wt_q4 wt_q5 wt_q6 wt_q7 wt_q8 wt_q9 wt_q12 dfi = sex lot wt90 breeding mc4r a mc4r d dfi_ql dfi_q6 dfi_q8 dfi_q11 dfi_q12 animal Classification of animals EXAMPLE 4: Conjugated gradient algorithms Given the entries A, b, a starting value x, a (perhaps implicitly defined) preconditioner M, a maximum number of repetitions imax and an error tolerance [epsilon] < 1 : silon] 2d0fDpt, an EXAMPLE 5: Arrangement of multiple markers (determination of informativeness).

Consider a chromosome fragment containing a quantitative trait site (QTL) and a group of markers (N1, N2, ..., Nn) on the left side of QTL and another group of markers (M ?, M2, ... , Mm) on the right side of QTL.

The present invention provides algorithms for detecting a group of informational flanking markers (N, M,) near QTL. This algorithm works in a similar way to the dimensionable window that moves around the chromosome fragment to locate a group of informational flanking markers, one is on the left side of QTL and the other on the right side of QTL. The following example illustrates that N¡ and M2 are a group of markers that are closer to QTL and informative (the link phase is known).

EXAMPLE 6 Pre-conditioning block-diagonal of variable size Solving the mixed model equations using the conjugate preconditioning gradient (PCCG) is the central part of MA-BLUP. The equations can be expressed in the matrix annotation assuming there are 6 animals involved: av, anana aiSaí6 l? A2la? 2Aa2i ° 26 «31« 32a33 «34a35a3í, (1) a ia? 2a-ßaYl 5a4b a5l« 51fíS3aS4- 5Sa $ 6. ß6J -ß63flfi fl65afi6 The diagonal elements (a11: a12, ..., a66) are most commonly used for pre-conditioning. The block diagonal of constant size such as It is recommended in the literature for pre-conditioning. In contrast, the methods and systems of the present invention provide for the use of a block diagonal of varying size such as 344 ¿? 45a46 ^ 22 ^ 3 a5 ü55as6 The size of each block diagonal is determined by the nature of the equations of the mixed model MA-BLUP.

Repetition In Data (lOD) Combined with PCCG Due to the nature of the mixed model equations, most of the elements in equation (1) above are zeros. MA-BLUP first processed the data and stored the non-zeros contributed from each data record for the equation of the mixed model on the hard disk. MA-BLUP currently does not build the elements a and s, in computer memory. Only store? S, b 's and block diagonals. Accordingly, the methods and systems of the present invention provide the algorithms that are repeated on each other data record, again and again until it converges.

EXAMPLE 7: Comparison of the analysis according to the present invention with the previously existing program, ISU-MABLUP The program of the State University of Iowa (ISU) is based on the public version of Matvec. The test was carried out by comparing the speed and efficiency of an MA-BLUP according to the present invention with the ISU package. Comparisons for speed are shown in the unit of either minute (m), hour (h), or day (d) when appropriate. 7. 1 Using ISU data groups ISU-MABLUP comes with its own sets of test data, which will be used to compare two packets. 7. 1.1 Small groups of data These are simulated data with 14 animals. The number of QTL features for each QTL model are shown below.

TABLE 7 Both ISU packets and the currently described invention generate the 'identical' results (indicated by '+') for one of the four previous QTL models. The meaning of 'identical' results has two folds (1) that refer only to a value of estimable function (2) and refers only to the first 4 digits after the decimal point.

TABLE 8 7. 1.2. Large groups of data There are two traits, two QTLs and 12,643 animals. Both ISU packets and the invention presently present generate the 'identical' results.

Use of large groups of data Two groups of data of approximately 63,000 animals were used. One group of data contains one QTL and another contains two QTLs. An extensive test was made and the comparison of the lOD solver was made as it is one of the most robust and efficient solvers available in the MABLUP analysis. Two platforms were used. These are the 32-bit Intel PC with Linux and a 64-bit Sparcstation group with Solaris (Computer Farm). All tests generated 'identical' results. The speed, however, was varied from platform to platform of individual trait to multiple trait. The comparisons for speed are shown in the following three tables. 7. 2.0.1 A QTL TABLE 9 7. 2.0.2. Two QTL TABLE 10 7. 2.0.3. No QTL In order to examine any differences in the polygenic effect resulting from the incorporation of QTL associated with the marker in the genetic evaluation system, MABLUP was run without QTL in the linear model. The group of data used is one containing a QTL.

TABLE 11 7. 3 Present invention against MTDFREML When using a group of different data comprising four traits and 28,624 animals. The comparison for the speed will then be in the minute unit (m). Note that the fastest solver (IOC_PCCG) was used in the aspect of the present invention.

TABLE 12 EXAMPLE 8: Change of the inbreeding coefficient for a QTL The conditional probability that two homologous alleles in the linked QTL marker (MQTL) at the individual site / are identical by ancestry, the given Gobs is defined as the inbreeding coefficient for a QTL; f¡ = Pr (Q1, = Q2 Gobs) This is different from Wright's inbreeding coefficient, which is the conditional probability that two homologous alleles at any site in / individual are identical in descent, given only one pedigree. The pair of two homologous alleles in MQTL, Q1¡ and Q2, - in / individual descended from one of the following pairs of parents: Let Tks d denote the event that the pair of alleles in i descends from the pair of parents * to ks.kcF 1 or 2. Now if / can be written as: Then Pr (Tkskdl 0bs) can be expressed in terms of probability of the descendant for the QTL alleles as, for example: where Bj (l, k) is the probability of the descendant for the allele QTL k to the allele /.

REFERENCES The following references, to the extent that they provide an illustrative procedure or other supplementary details to those set forth herein, are specifically incorporated by reference. Abdel-Azim G. and A. E. Freeman. 2001. A rapid method for computing the inverse of the gametic covariance matrix between relatives for a marked quantitative trait locus. Genet Sel. Evol., 33: 153-173. Chakraborty, R., Moreau, L., Dekkers, J.C. 2002. A method to optimize selection on multiple identified quantitative trait loci. Genet Sel. Evol. 34 (2): 145-70. Falconer, DS and Mackay, Introduction to Quantitative Genetics, TF C, Eds., Longman Group Limited, Longman House, Burnt Mill, Harlow Essex 2JE, England.4th Edition, 1986. Fernando, RL and Grossman, M. 1989. "Marker assisted selection using best linear unbiased prediction, "Genet. Sel. Evol. 21: 467-477. Gibson, J. P. 1994. Short-term gain at the expense of long-term response with selection of identified loci. Proceedings of the 5th World Congress on Genetics Applied to Livestock Production, Guelph, 21: 201-204. * Henderson, C. R. 1984. Applications of Linear Models in Animal Breeding. Published by the University of Guelph, Guelph, Ontario, Canada. Hernandez-Sanchez, J., Visscher, P., Plastow, G. and Haley, C. 2003. Candidate Gene Analysis for Quantitative Traits Using the Transmission Disequilibrium Test: The Example of the Melanocortin 4-Receptor in Pigs. Genetics 164: 637-644. Kim, K. S., Larsen, N., Short., T., Plastow, G. and Rothschild, M. F. 2000. A missense variant of the porcine melanocortin-4 receptor (MC4R) gene is associated with fatness, growth, and feed intake traits. Mammalian Genome. 11: 131-135. Lidauer, M., Stnden, I., Mantysaari, E.A., Poso, J., and A. Kettunen. 1999, "Solving Large Test-day Models by Teaching on Data and Preconditioned Conjungate Gradient," J. Dairy Sci. 82: 2788-2796. Mal cot, G., 1948 Les Mathematiques de l'Heredite. Masson, Paris.) Milan, D., ei al. 2000. "A mutation in PRKAG3 associated with excess glycogen content in pig skeletal muscle." Science, 288: 1248-1251, Pong-Wong, R., George, AW, Woolliams, JA, and CS Haley, 2001. "A simple and rapid method for calculating identity-by-descent matrices using multiplemarkers, "Genet, Sel. Evolution 33: 453-471, Quaas, RL, Anderson, RD, Gilmour, AR, 1984. BLUP school handbook; Use of mixed models for prediction and estimation of (co) variance components, Animal Breeding and Genetics Unit, University of New England, NSW 2351, Australia, Strandén, I. and M. Lidauer, 1999. "Solving large mixed linear models using preconditioned conjugate gradient iteration," J Dairy Sci. 88: 2779-2787. Shewchuk, JR 1994"An introduction to the conjugate gradient method without the agonizing pain. Tech. Rep. CMU-CS-94-125, Carnegie Mellon University, Pittsburgh, Pennsylvania. Totir, L.R. 2002. Genetic evaluation with finite locus models. PhD Dissertation. Lowa State University, Ames, Iowa. Tsuruta, S., Misztal, I., and I. Stranden. 2001. "Use of the preconditioned conjugate gradient algorithm as a generic solver for mixed-model equations in animal breeding applications," J. Animal Sci. 79: 1166-1172. Van Vleck, LD, Pollak, EJ, and Oltenacu, EAB, Genetics for the Animal Sciences, WH Freeman and Company, New York, 1987 Wang, T., Fernando, RL, van der Beek, S., Grossman, M., and JAM van Arendonk. 1995. "Covariance between relatives for a marked quantitative trait locus." Genet Sel. Evol. 27: 251-274 Wang, T., Fernando, R.L., Stricker, C. and R.C. Elston. 1996"An approximation to the likelihood for a pedigree with loops." Theor. Appl. Genet 93: 1299-1309. WO 02/20850 A2, Rothschild et al., March 14, 2002.

Claims (34)

NOVELTY OF THE INVENTION CLAIMS

1. - A method to increase an average genetic importance in the animal population, which includes: a. Select one or more traits for which an improved generic importance is desired; b. Select one or more quantitative trait sites (QTL) for each selected trait; c. Select three or more molecular genetic markers of interest for each QTL for each selected trait, d. Provide databases that include: i. genotype data for three or more molecular genetic markers for each selected trait, for a plurality of animals in the population; ii. provide pedigree data for each animal in the population; iii. optionally, data for one or more fixed effects; and. Use a computer program capable of carrying out the best linear unbiased assisted marker prediction to simultaneously analyze the data from the provided databases to calculate a classification of animals; wherein the computer program uses a diagonal variable block preconditioned gradient (PCCG) algorithm to classify the animals; where the animals are classified according to their estimated breeding value (EBV) for the selected molecular genetic markers and, if provided, the quantitative traits.

2. The method according to claim 1, further characterized in that it comprises the use of EBVs calculated to prepare a breeding plan for the animal population that provides the optimum improvement in the genetic importance of the population.

3. The method according to claim 1, further characterized in that the population of animals is a herd of pigs.

4. The method according to claim 1, further characterized in that the trait is selected from the group consisting of: efficient growth traits, meat quality traits, breeding traits, and health traits.

5. The method according to claim 1, further characterized in that the molecular genetic markers are selected from any known polymorphism that affects the expression of mRNA or protein of a gene.

6. The method according to claim 5, further characterized in that it is selected from the group consisting of: single nucleotide polymorphisms, single sequence repeats, protein point mutations, and gene isoforms.

The method according to claim 3, further characterized in that at least one molecular genetic marker is selected from those markers that are known to modulate a favorable phenotype.

The method according to claim 3, further characterized in that at least one of the molecular genetic markers is a marker selected from the group consisting of: an individual nucleotide polymorphism in the porcine PRKAG3 gene (protein kinase, subunit range 3 activated by AMP), and a polymorphism in the porcine melanocortin-4 receptor.

The method according to claim 3, further characterized in that at least one of the molecular genetic markers is a marker for an individual nucleotide polymorphism in the porcine PRKAG3 gene.

10. The method according to claim 1, further characterized in that the computer program uses a repeating algorithm on data (lOD).

The method according to claim 1, further characterized in that the output of the computer program further comprises results indicating the informativity of one or more of the molecular genetic markers selected for at least one quantitative trait site (QTL) and / or a calculation of the genetic proximity / proximity of one or more molecular markers to at least one QTL.

The method according to claim 11, further characterized in that the molecular genetic markers have the highest degree of informativity and / or approximation for at least one QTL are identified.

13. The method according to claim 1, further characterized in that the computer program uses a programming feature to improve the ease of the user interface.

The method according to claim 1, further characterized in that the selected molecular genetic markers comprise a marker haplotype.

15. A system to increase the average genetic significance of the animal population for one or more selected traits, the system comprises: a. a computer; b. a computer-accessible database that provides data in one or more quantitative data sites (QTL) for each selected trait; c. a computer-accessible database that provides data, for animals in the population, for three or more molecular genetic markers for each QTL selected for each selected trait; d. a computer-accessible database that provides pedigree data for animals in the population; and. optionally a computer-accessible database that provides individual data for each animal in the population for at least one fixed effect; F. a computer program capable of carrying out the best linear unbiased prediction assisted by marker and simultaneously evaluating the data in all the databases and the classification of animals in the population according to their respective estimated breeding value for each of the selected features; wherein the computer program uses a diagonal variable block preconditioned gradient (PCCG) algorithm to classify the animals; g. a user interface that includes a data entry system, said user interface coupled to said computer and configured to allow the user to instruct the computer to access the available bases and use the computer program to generate the output that includes a classification of the animals according to their estimated breeding values and / or their individual estimated breeding values.

16. The system according to claim 15, further characterized in that the animal population is a herd of pigs.

The system according to claim 16, further characterized in that at least one of the molecular genetic markers is selected from the group consisting of markers for the porcine PRKAG3 gene and the gene encoding the melanocortin-4 receptor.

18. The system according to claim 16, further characterized in that at least one of the molecular genetic markers is a marker for an individual nucleotide polymorphism in the porcine PRKAG3 gene.

19. The system according to claim 16, further characterized in that the selected molecular genetic markers comprise a marker haplotype.

20. A system to identify the molecular genetic marker (s) that has the highest degree of informativity for one or more selected quantitative trait sites (QTL), the system comprises: a. a computer; b. a computer-accessible database that provides individual data, for animals in the population, for three or more molecular genetic markers for each selected quantitative trait site; c. a computer program capable of simultaneously evaluating the data in all the databases and determining the relative informativity for each of the molecular genetic markers for which the data are provided; wherein the computer program is capable of carrying out the best unbiased linear prediction assisted by marker and uses a diagonal variable block preconditioned gradient algorithm (PCCG) to determine the relative informativity of each molecular genetic marker; d. a user interface that includes a data entry system, said user interface coupled to said computer and configured to allow the user to instruct the computer to access the available database and use the computer program to generate the output that includes an indication of the informativity of each molecular genetic marker for which the data are provided.

The system according to claim 20, further characterized in that the quantitative trait site is selected from any known site that is associated with a known trait.

22. The system according to claim 20, further characterized in that the quantitative trait site is selected from any trait site selected from the group consisting of efficient growth traits, meat quality traits, breeding traits, and traits of Health.

The system according to claim 20, further characterized by providing a computer-accessible database containing individual data for animals in the population for at least one fixed effect; wherein the computer executable program is able to simultaneously evaluate the data in all provided databases and classify the animals in the population according to their respective estimated breeding value for each of the selected traits.

24. The system according to claim 20, further characterized in that the selected molecular genetic markers comprise a marker haplotype.

25. The method according to claim 1, further characterized in that it comprises the use of animal classifications to identify the optimal breeding pairs in the population.

26. The method according to claim 25, further characterized in that the molecular genetic markers selected comprise a marker haplotype.

27. A method to improve one or more meat quality traits in pigs, the method comprises: a. classifying a plurality of pigs to identify the nature of one or more individual nucleotide polymorphisms (SNPs) in the porcine PRKAG3 gene, is selected from the group consisting: an A / G at position 51, an A / G in position 462, an AG in position 1011, a C / T in position 1053, a C / T in position 2475, an A / G in position 2607, an A / G in position 2906, an A / G at position 2994, and a C / T at position 4506, where all the numbering is according to the sequence SEQ ID NO: 1 and identifying those having a desired allele; b. select those pigs identified as having a desired allele; c. use the selected pigs as offspring / females in the breeding plan to produce progeny; where the progeny has an increasing frequency of the desired allele.

28. The method according to claim 27, further characterized in that the presence or absence of the polymorphism is determined through a selected method consisting of: DNA sequencing, restriction fragment length polymorphism analysis (RFLP); heteroduplex analysis, conformational polymorphism analysis of individual chain structure (SSCP), denaturation gradient gel electrophoresis (DGGE), real-time PCR analysis (TAQMAN®), temperature gradient gel electrophoresis (TGGE), extension of primer, allele-specific hybridization, and INVADER® genetic analysis assay.

29. The method according to claim 27, further characterized in that at least one quality trait of meat is selected from the group consisting of increased pH, and a decreased 7-day depuration.

30. A team to detect the nature of one or more polymorphisms in the PRKAG3 gene; the equipment comprises means for detecting the polymorphism in DNA, and RNA of the gene; wherein the polymorphisms are selected from the group consisting of one or more of the following SNP (s): an A / G at position 51, an A / G at position 462, an A / G at position 1011, a C / T in position 1053, a C / T in position 2475, an A / G in position 2607, an A / G in position 2906, an A / G in position 2994, and a C / T in position 4506, where all the numbering is in accordance with the sequence SEC lD NO: 1.

31. The equipment according to claim 30, further characterized in that while the polymorphism is detected through one or more of the following detection means: DNA sequencing, restriction fragment length polymorphism analysis (RFLP), analysis of heteroduplex, conformational polymorphism of individual chain structure (SSCP), denaturation gradient gel electrophoresis (DGGE), polymerase chain reaction (PCR), real-time PCR analysis (TAQMAN®), gradient gel electrophoresis temperature (TGGE), enzyme-linked immunosorbent assay (ELISA) and other immunoassays; wherein the equipment comprises one or more of the following; restriction endonuclease enzyme, DNA polymerase, a reverse transcriptase, a pH regulator, deoxyribonucleotides, an oligonucleotide suitable for use as a DNA or RNA probe, an oligonucleotide suitable for use as an initiator in the synthesis of DNA or RNA, a fluorescent label, and an antibody.

32. An oligonucleotide suitable for use in an equipment according to claim 31.

33. The oligonucleotide according to claim 32, further characterized in that it is selected from primers comprising the sequence of any of the primers listed in Table 1 (SEQ. NO.-2-17).

34. The oligonucleotide according to claim 32, further characterized in that it is selected from the group consisting of the primers provided in Table 1 (SEQ ID NO: 2-17).