Difference between revisions of "Genomic Evaluation Guidelines"

From BIF Guidelines Wiki
 
(33 intermediate revisions by 2 users not shown)
Line 1: Line 1:
DNA technology has developed rapidly in the past decade and now has a variety of
+
[[Category: Genetic Evaluation]]
 +
[[Category:Data Collection]]
 +
DNA technology has developed rapidly in the past nearly two decades and now has a variety of
 
applications. For beef cattle genetic improvement, the primary areas of application are
 
applications. For beef cattle genetic improvement, the primary areas of application are
 
pedigree validation, parentage determination, and gene-based (genotypic) selection.
 
pedigree validation, parentage determination, and gene-based (genotypic) selection.
Individual and parentage verification are now routine practices, while gene-based
+
Individual and parentage verifications are now routine practices, and genomic data are now routinely used in [[Expected Progeny Difference | EPD]] production. This chapter describes the current uses of
selection is in the early stages of development. This chapter describes current uses of
 
 
DNA technology and provides an overview of applications currently under development.
 
DNA technology and provides an overview of applications currently under development.
==Types of DNA Markers==
+
==DNA Markers==
 
Analytical techniques to differentiate DNA of individuals or populations require genetic
 
Analytical techniques to differentiate DNA of individuals or populations require genetic
 
markers, which are defined as identifiable DNA segments that differ in nucleotide
 
markers, which are defined as identifiable DNA segments that differ in nucleotide
sequence from one individual to the next. Two types of markers may be used:
+
sequence from one individual to the next. The current standard for identifying genomic markers is single nucleotide polymorphisms (SNPs). [https://www.illumina.com/science/technology/beadarray-technology.html Beadarray based "chips"] create uniquely
microsatellites and single nucleotide polymorphisms (SNPs). Both create uniquely
 
 
identifiable DNA patterns that may be used to follow the transmission of specific
 
identifiable DNA patterns that may be used to follow the transmission of specific
 
chromosomal regions from parents to progeny.
 
chromosomal regions from parents to progeny.
  
Microsatellite markers are segments of chromosomal DNA that include a variable
+
As the name implies, SNPs are a change (mutation) from the specific nucleotide originally present in a
number of repeated two to six nucleotide base sequences. Such markers are
 
interspersed throughout the genome and are generally found in non-coding regions.
 
These repetitive regions are subject to additions and subtractions in the number of
 
tandem repeats of basic two to six nucleotide segments, and this creates uniquely
 
identifiable alleles at each site within the genome where the particular microsatellite is
 
found. Microsatellites routinely have been used in parentage analysis, because
 
multiple alleles generally found at each locus make them highly informative. They have
 
provided the basis for individual and parentage identification in humans, dogs, cattle,
 
and many other species.
 
 
 
Single nucleotide polymorphisms are the type of other marker. As the name implies,
 
they are a change (mutation) from the specific nucleotide originally present in a
 
 
particular location in an individual to a different nucleotide at that same site and are
 
particular location in an individual to a different nucleotide at that same site and are
transmitted from parent to offspring, just like any other gene. Across evolutionary time,
+
transmitted from parent to offspring, just like a gene. Across evolutionary time,
 
thousands of SNPs have been created by mutation. They now can be found every 100
 
thousands of SNPs have been created by mutation. They now can be found every 100
to 300 bases throughout the 3 billion base pairs in the genome. Because SNPs are
+
to 300 bases throughout the 3-billion base pairs in the genome. Because SNPs are
 
widely distributed, it is likely that any gene of economic importance is located closely
 
widely distributed, it is likely that any gene of economic importance is located closely
 
adjacent to several SNPs that can be used to mark its presence.
 
adjacent to several SNPs that can be used to mark its presence.
SNP markers promise to be increasingly useful in the future for developing high-
+
SNP markers promise to be increasingly useful in the future for developing high-resolution maps because of their high throughput capability and potentially low cost in generating data.
resolution maps because of their high throughput capability and potentially low cost.
+
With the availability of whole-genome sequences, SNPs that are dispersed across all
With the availability of whole genome sequences, SNPs that are dispersed across all
+
chromosomes present important advantages as markers for genomic analysis.
chromosomes, present important advantages as markers for genome analysis.
+
 
 
Some SNPs are located within the coding region of a gene and can affect the structure
 
Some SNPs are located within the coding region of a gene and can affect the structure
 
and function of a protein. This type of variation may be directly responsible for
 
and function of a protein. This type of variation may be directly responsible for
Line 41: Line 29:
 
Other SNPs occur either “upstream” or “downstream” of the coding region of a gene and
 
Other SNPs occur either “upstream” or “downstream” of the coding region of a gene and
 
may influence the regulation of gene expression. Others occur in locations that do not
 
may influence the regulation of gene expression. Others occur in locations that do not
interfere with the structure or production of a protein. SNPs have the advantage that
+
interfere with the structure or production of a protein.
they are less likely to undergo spontaneous mutation than microsatellites; thus they are
 
inherited with greater stability.
 
  
 
==DNA Collection==
 
==DNA Collection==
Line 49: Line 35:
 
muscle, fat, white blood cells found in blood and milk, skin, and epithelial cells collected
 
muscle, fat, white blood cells found in blood and milk, skin, and epithelial cells collected
 
from saliva. Minute amounts of tissue, such as a single drop of blood or several
 
from saliva. Minute amounts of tissue, such as a single drop of blood or several
mucosal cells, are all that are required for routine DNA analysis. Common collection
+
mucosal cells, are all that are required for routine DNA analysis. [[Genotyping |Common collection
methods include a drop of blood blotted on a paper that is dried, covered, and stored at
+
methods]] include a drop of blood blotted on a paper that is dried, covered, and stored at
 
room temperature, ear tag systems that deposit a tissue sample in an enclosed
 
room temperature, ear tag systems that deposit a tissue sample in an enclosed
 
container with bar code identification, and hair follicles. Techniques have been
 
container with bar code identification, and hair follicles. Techniques have been
 
developed that allow for rapid release of DNA from cells and immediate analysis of the
 
developed that allow for rapid release of DNA from cells and immediate analysis of the
 
samples.
 
samples.
 +
 
==Combining Molecular and Quantitative Approaches in Genetic Evaluation==
 
==Combining Molecular and Quantitative Approaches in Genetic Evaluation==
Research into the molecular basis of inheritance is progressing at a rapid pace.
+
 
Technologies that permit identification of molecular genetic differences in
 
deoxyribonucleic acid (DNA) sequence among animals are also evolving very rapidly.
 
Several DNA-based tools are being marketed in the beef industry; some as selection
 
tools. These tools are known by a variety of names in the academic community and
 
within the beef industry (e.g., genomic tests, DNA markers, molecular tests or markers).
 
DNA-based selection tools present opportunities and challenges to the U.S. beef
 
industry. Accurate DNA-based selection tools will give beef cattle breeders opportunity
 
to identify animals with superior breeding value (BV) as soon as a tissue sample can be
 
collected and analyzed, potentially leading to significant savings in time and money
 
associated with performance testing and genetic evaluation. However, as currently
 
marketed, the BV information provided by DNA-based tools is not uniformly reported
 
and the proportion of variation in true BV accounted for by the tools is unknown.
 
Further, the BV information provided by competing DNA-based tools overlaps and is not
 
independent of information provided by current national cattle evaluation (NCE)
 
systems.
 
 
Performance testing and genetic evaluation are being conducted on an increasing
 
Performance testing and genetic evaluation are being conducted on an increasing
 
number of traits. Types of information available (i.e., available from a practical and
 
number of traits. Types of information available (i.e., available from a practical and
 
economical view) vary among traits. Types of information include pedigree
 
economical view) vary among traits. Types of information include pedigree
 
relationships, performance measurements (i.e., phenotypes), and DNA test results.
 
relationships, performance measurements (i.e., phenotypes), and DNA test results.
Phenotypes may include direct and indirect measurements on the same trait. Table 1
+
Phenotypes may include direct and indirect measurements on the same trait.  
illustrates various combinations. Because most animals marketed in the U.S. as
+
 
seedstock have known parentage the table assumes that pedigree relationships are
+
Some traits are difficult to measure for which there are currently no DNA tests available. These
known.
 
Some traits are difficult to measure for which there are no DNA tests available. These
 
 
traits will likely be the focus of future research. In a second category are traits for which
 
traits will likely be the focus of future research. In a second category are traits for which
 
phenotypes are regularly measured in the field, systematically data-based, and for
 
phenotypes are regularly measured in the field, systematically data-based, and for
Line 87: Line 57:
 
current example would be tenderness. Tenderness phenotypes are difficult and
 
current example would be tenderness. Tenderness phenotypes are difficult and
 
expensive to measure, but DNA tests are available. In a fourth category are traits where
 
expensive to measure, but DNA tests are available. In a fourth category are traits where
both phenotypes and DNA tests are available. A current example would be carcass
+
both phenotypes and DNA tests are available. A current example would be [[Marbling score | carcass
marbling
+
marbling]].
Table 1. Traits categorized according to information available.
+
 
Industry-collected
+
===Guiding Philosophy===
Phenotypes
+
''BIF believes that information from DNA tests only has value in selection when incorporated with all other available forms of performance information for economically important traits in NCE, and when communicated in the form of an EPD with corresponding BIF accuracy. For some economically important traits information other than DNA tests may not be available. Selection tools based on these tests should still be expressed as EPD within the normal parameters of NCE.''
DNA Tests
+
 
No
+
==Validation==
Yes
+
DNA tests are developed based on associations between variations in base-pair
No Yes
+
sequences at one or more loci with variations in phenotypes. The animal populations
---
+
used to develop the test may or may not be representative of beef industry populations.
EPD EPD
+
Validation generally involves the confirmation or rejection of these associations in
EPD
+
populations that are representative of the beef industry and different from those in which
1
+
the tests were developed. Validation studies are considered to be more reliable if
Prepared by M. W. Tess and the BIF Commission on DNA Markers. Commission members:
+
conducted by scientists who have no vested interest in the tests (e.g., development,
Bill Bowman, Ronnie Green, Ronnie Silcox, Darrell Wilkes, and Jim Wilton.
+
commercialization, or marketing). To date, components of commercially available DNA
+
tests have been validated by the National Beef Cattle Evaluation Consortium (NBCEC)
==Guiding Philosophy==
+
serving as an independent third party. Validation serves to reduce risk for breeders
BIF believes that information from DNA tests only has value in selection when incorporated with
+
using DNA tests for selection.
all other available forms of performance information for economically important traits in NCE,
+
 
and when communicated in the form of an EPD with corresponding BIF accuracy. For some
+
''BIF recommends that breeders who use DNA tests should, whenever possible, choose DNA tests that have been validated in populations that are representative of the beef cattle industry by scientists independent of the organization that developed or will market the test.''
economically important traits information other than DNA tests may not be available. Selection
+
 
tools based on these tests should still be expressed as EPD within the normal parameters of
+
==Assessment==
NCE.
+
Assessment involves determining how specific DNA tests are associated with each
 +
other and with non-target phenotypes. Assessment seeks to determine how competing
 +
DNA tests overlap and how non-target traits will be influenced by selection based on
 +
these tests. For example, it is important to know if selection based on a DNA test for
 +
[[Measures of Tenderness | tenderness]] has any desirable or adverse effects on other economically important traits
 +
(growth, feed intake, fertility, etc.). As with validation, assessment studies are
 +
considered to be more reliable if conducted by scientists who have no vested interest in
 +
the tests.
 +
 
 +
''BIF recommends that assessment studies should be conducted in populations that are representative of the beef cattle industry by scientists independent of the organization that developed or will market the test.''
 +
 
 +
==Inclusion of DNA Test Information in NCE Programs==
 +
Statistical procedures for incorporating DNA test information into NCE and the
 +
computation of EPD and associated accuracies are described in [[Single-step Hybrid Marker Effects Models]] and [[Single-step Genomic BLUP]]. Results of the evaluation of a DNA test will also provide estimated genetic correlations
 +
among competing DNA tests, genetic correlations between DNA tests and non-target
 +
traits, and the fraction of the additive genetic variance of the target trait accounted for by
 +
the DNA test.
 +
 
 +
Results of the evaluation phase (described above) will provide all the statistical
 +
parameters needed for NCE. The decision to include a DNA test in a NCE system
 +
should be made by the organization responsible for computing the EPD. Consideration
 +
should be given to heritability of the trait, availability of producer-collected phenotypes,
 +
and increase in accuracy provided by the addition of the DNA test information.
 +
BIF recommends that a DNA test should be considered for inclusion in the NCE system
 +
when, after estimating the covariances and running the NCE system, use of the DNA
 +
test results in more accurate EPD at a young age.
 +
 
 +
Reporting of DNA Test Results by Genomic Companies
 +
It is important the DNA test results be reported to the beef industry in a consistent,
 +
understandable format. Further, the format should be compatible with NCE methods. It’s
 +
possible that a single DNA test (i.e., genotypes from a single panel of markers) may
 +
yield information useful for both management and selection. Predictors based on these
 +
tests should be clearly identified with respect to their uses – i.e., future phenotypes
 +
versus breeding value.
 +
 
 +
''BIF recommends that DNA test results be reported in the form of an EPD, in units of the trait, on a continuous scale, and with a corresponding BIF accuracy. It is likely that research will develop new DNA tests for traits that have no industry-collected phenotypes. If the target trait is measured in the reference populations, evaluation of the DNA test as a selection tool should be as described above.''
 +
 
 +
==Novel Traits==
 +
It’s conceivable that the target traits for some new DNA tests may not be measured in
 +
reference populations. In such cases precise definition of the target trait will be
 +
important.
 +
 
 +
An independent organization such as NBCEC should conduct or coordinate the
 +
validation studies of DNA tests for novel traits. Validation may be approximated by review and
 +
(or) re-analysis of data used to develop the test. Such data should include DNA test
 +
results, phenotypes, and pedigree relationships. Data used to develop such new tests
 +
should be of sufficient quality and quantity to allow the estimation of the additive genetic
 +
variance of the target trait and the covariance between the DNA test score and the
 +
target trait.
 +
 
 +
''BIF recommends that, for DNA tests targeting traits that have no industry-collected phenotypes and for which no phenotypes are collected in reference populations, results should be reported in the form of an EPD, in the units of the trait, on a continuous scale, and with a corresponding BIF accuracy.''
 +
 
 +
==Attribution==
 +
Information in this article was derived from Chapter 4 of the 9th edition of the BIF Guidelines, with substantial modification for updating to current technology and methods. The attribution in that chapter was to: M. W. Tess and the BIF Commission on DNA Markers. Commission members:
 +
Bill Bowman, Ronnie Green, Ronnie Silcox, Darrell Wilkes, and Jim Wilton. Please view the history link for this article for more information on these modifications.

Latest revision as of 18:28, 12 April 2021

DNA technology has developed rapidly in the past nearly two decades and now has a variety of applications. For beef cattle genetic improvement, the primary areas of application are pedigree validation, parentage determination, and gene-based (genotypic) selection. Individual and parentage verifications are now routine practices, and genomic data are now routinely used in EPD production. This chapter describes the current uses of DNA technology and provides an overview of applications currently under development.

DNA Markers

Analytical techniques to differentiate DNA of individuals or populations require genetic markers, which are defined as identifiable DNA segments that differ in nucleotide sequence from one individual to the next. The current standard for identifying genomic markers is single nucleotide polymorphisms (SNPs). Beadarray based "chips" create uniquely identifiable DNA patterns that may be used to follow the transmission of specific chromosomal regions from parents to progeny.

As the name implies, SNPs are a change (mutation) from the specific nucleotide originally present in a particular location in an individual to a different nucleotide at that same site and are transmitted from parent to offspring, just like a gene. Across evolutionary time, thousands of SNPs have been created by mutation. They now can be found every 100 to 300 bases throughout the 3-billion base pairs in the genome. Because SNPs are widely distributed, it is likely that any gene of economic importance is located closely adjacent to several SNPs that can be used to mark its presence. SNP markers promise to be increasingly useful in the future for developing high-resolution maps because of their high throughput capability and potentially low cost in generating data. With the availability of whole-genome sequences, SNPs that are dispersed across all chromosomes present important advantages as markers for genomic analysis.

Some SNPs are located within the coding region of a gene and can affect the structure and function of a protein. This type of variation may be directly responsible for differences among individuals in phenotypic merit for economically important traits. Other SNPs occur either “upstream” or “downstream” of the coding region of a gene and may influence the regulation of gene expression. Others occur in locations that do not interfere with the structure or production of a protein.

DNA Collection

DNA is found in every nucleated cell in the body. It can be extracted from semen, muscle, fat, white blood cells found in blood and milk, skin, and epithelial cells collected from saliva. Minute amounts of tissue, such as a single drop of blood or several mucosal cells, are all that are required for routine DNA analysis. Common collection methods include a drop of blood blotted on a paper that is dried, covered, and stored at room temperature, ear tag systems that deposit a tissue sample in an enclosed container with bar code identification, and hair follicles. Techniques have been developed that allow for rapid release of DNA from cells and immediate analysis of the samples.

Combining Molecular and Quantitative Approaches in Genetic Evaluation

Performance testing and genetic evaluation are being conducted on an increasing number of traits. Types of information available (i.e., available from a practical and economical view) vary among traits. Types of information include pedigree relationships, performance measurements (i.e., phenotypes), and DNA test results. Phenotypes may include direct and indirect measurements on the same trait.

Some traits are difficult to measure for which there are currently no DNA tests available. These traits will likely be the focus of future research. In a second category are traits for which phenotypes are regularly measured in the field, systematically data-based, and for which EPDs are computed. The emergence of DNA tests now permits estimation of BV on animals for which little or no phenotypic information is available (a third category). A current example would be tenderness. Tenderness phenotypes are difficult and expensive to measure, but DNA tests are available. In a fourth category are traits where both phenotypes and DNA tests are available. A current example would be carcass marbling.

Guiding Philosophy

BIF believes that information from DNA tests only has value in selection when incorporated with all other available forms of performance information for economically important traits in NCE, and when communicated in the form of an EPD with corresponding BIF accuracy. For some economically important traits information other than DNA tests may not be available. Selection tools based on these tests should still be expressed as EPD within the normal parameters of NCE.

Validation

DNA tests are developed based on associations between variations in base-pair sequences at one or more loci with variations in phenotypes. The animal populations used to develop the test may or may not be representative of beef industry populations. Validation generally involves the confirmation or rejection of these associations in populations that are representative of the beef industry and different from those in which the tests were developed. Validation studies are considered to be more reliable if conducted by scientists who have no vested interest in the tests (e.g., development, commercialization, or marketing). To date, components of commercially available DNA tests have been validated by the National Beef Cattle Evaluation Consortium (NBCEC) serving as an independent third party. Validation serves to reduce risk for breeders using DNA tests for selection.

BIF recommends that breeders who use DNA tests should, whenever possible, choose DNA tests that have been validated in populations that are representative of the beef cattle industry by scientists independent of the organization that developed or will market the test.

Assessment

Assessment involves determining how specific DNA tests are associated with each other and with non-target phenotypes. Assessment seeks to determine how competing DNA tests overlap and how non-target traits will be influenced by selection based on these tests. For example, it is important to know if selection based on a DNA test for tenderness has any desirable or adverse effects on other economically important traits (growth, feed intake, fertility, etc.). As with validation, assessment studies are considered to be more reliable if conducted by scientists who have no vested interest in the tests.

BIF recommends that assessment studies should be conducted in populations that are representative of the beef cattle industry by scientists independent of the organization that developed or will market the test.

Inclusion of DNA Test Information in NCE Programs

Statistical procedures for incorporating DNA test information into NCE and the computation of EPD and associated accuracies are described in Single-step Hybrid Marker Effects Models and Single-step Genomic BLUP. Results of the evaluation of a DNA test will also provide estimated genetic correlations among competing DNA tests, genetic correlations between DNA tests and non-target traits, and the fraction of the additive genetic variance of the target trait accounted for by the DNA test.

Results of the evaluation phase (described above) will provide all the statistical parameters needed for NCE. The decision to include a DNA test in a NCE system should be made by the organization responsible for computing the EPD. Consideration should be given to heritability of the trait, availability of producer-collected phenotypes, and increase in accuracy provided by the addition of the DNA test information. BIF recommends that a DNA test should be considered for inclusion in the NCE system when, after estimating the covariances and running the NCE system, use of the DNA test results in more accurate EPD at a young age.

Reporting of DNA Test Results by Genomic Companies It is important the DNA test results be reported to the beef industry in a consistent, understandable format. Further, the format should be compatible with NCE methods. It’s possible that a single DNA test (i.e., genotypes from a single panel of markers) may yield information useful for both management and selection. Predictors based on these tests should be clearly identified with respect to their uses – i.e., future phenotypes versus breeding value.

BIF recommends that DNA test results be reported in the form of an EPD, in units of the trait, on a continuous scale, and with a corresponding BIF accuracy. It is likely that research will develop new DNA tests for traits that have no industry-collected phenotypes. If the target trait is measured in the reference populations, evaluation of the DNA test as a selection tool should be as described above.

Novel Traits

It’s conceivable that the target traits for some new DNA tests may not be measured in reference populations. In such cases precise definition of the target trait will be important.

An independent organization such as NBCEC should conduct or coordinate the validation studies of DNA tests for novel traits. Validation may be approximated by review and (or) re-analysis of data used to develop the test. Such data should include DNA test results, phenotypes, and pedigree relationships. Data used to develop such new tests should be of sufficient quality and quantity to allow the estimation of the additive genetic variance of the target trait and the covariance between the DNA test score and the target trait.

BIF recommends that, for DNA tests targeting traits that have no industry-collected phenotypes and for which no phenotypes are collected in reference populations, results should be reported in the form of an EPD, in the units of the trait, on a continuous scale, and with a corresponding BIF accuracy.

Attribution

Information in this article was derived from Chapter 4 of the 9th edition of the BIF Guidelines, with substantial modification for updating to current technology and methods. The attribution in that chapter was to: M. W. Tess and the BIF Commission on DNA Markers. Commission members: Bill Bowman, Ronnie Green, Ronnie Silcox, Darrell Wilkes, and Jim Wilton. Please view the history link for this article for more information on these modifications.