​Livestock breeders have now realized that highly trait-specific selective breeding might improve one feature to the detriment of others that are critical to animal health. 

In milk cows,

But what we want to know is how, despite significant inbreeding depression, breeders were able to achieve such remarkable improvements in production traits? ​

In fact, commercial breeders do in fact choose the best offspring of each generation to breed to, but not just the top few animals. They select all of the best performing animals, perhaps 10-20% of the best animals. The goal is to capture the genetic variation present in the best animals. They know thatt each animal is a mix of genes from two good quality parents, and from the same parents this mix will be more fortuitous in some animals than others. The result is a collection of offspring that carry the genes of good producers, and when bred together in the next generation the new mixes of genes will once again produce some animals that are better than others. By selecting many of the top animals in each generation, the gene pool of the popualtion is gradually shifted in the direction of higher productivity as each new generation produces animals with a different mix of genes.
​

This assumes that genes are at least partly responsible for the traits of interest. If not, or if the influence is very small, you will not see imprrovement in performance with seletive breeding. But as in this example of selection for a trait like speed, selection of the 10% fastest animals in each generation will move the average performance of the animals in the population to higher speed. 

Of course, not only must the trait have some genetic basis, but for this to work you need enough genetic variation to produce many new combinations of genes in the next generation of offspring, some of which can perform better than their parents. The key here is genetic varaiation. If you only keep the top animal in each generation, you will lose the genetic variation needed to improve anything. The genes associated with the best performance will become fixed in the population, ending your genetic improvement program.

​Inbreeding and strong selection eliminates the raw material - genetic variation - that is necessary for improvement of traits in animals. Because of Inbreeding in each generation of dogs, genetic variation is lost and homozygosity increases, which can drive traits to unwanted extremes. Most breeders realize that breeding together two outstanding but closely related dogs doesn't guaranted outstanding puppies. This is because loci that were heterozygous in the parents can be homozygous in the offspring, which then lose the genetic advantage of heterozygosity. This is called "overdominance", where the phenotype of the heterozygous combination is superior to the phenotypes of either allele when homozygous. For traits that depend on overdominance for the best phenotype, inbreeding will destroy the advantage of heterozygosity. 

What's the lesson here?

One of the "pearls of wisdom" often offered by long-time breeders is that you should "breed the best to the best". This suggests that the secret to success is the consistent application of this simple rule of thumb. But what we know about genetics makes clear that, while this might produce nice puppies from a pair of parents, selecting the pick of that litter and doing the same in the next generation will leave a trail discarded genetic variation that might include the raw material you need to produce something better than either parents. It's the shuffling of this variation in each generation that provides the opportunity for fortuitious combinations that can create animals that are superior to their ancestors, generation after generation. This is how we got those massive turkeys and the amazing milkers. 

Dog breeders, the value of your dogs is in their genes; it's your money in the bank that will pay dividends generation after generation. Don't toss out the lesser dogs that happened to get the perfect mix of the available variation. 

DogsArk is designed to assist dog breeders in producing genetically sound puppies while preserving breed genetic diversity. It uses essential genetic concepts, including inbreeding, heterozygosity, kinship, and fixation index, to provide insights for informed breeding decisions.

The tool is structured into three modules:

1. Breed Summary (Module A)
Overview: Offers a summary of the breed's genetic diversity, genetic disorders, and traits.

Features:

• Dog Inventory: Provides a database of dogs, distinguishing between anonymous and known dogs.
• Genetic Diversity: Summarizes inbreeding coefficients, kinship, and heterozygosity with values for each measure.
• Genotype Frequencies: Reports health-related gene statuses and frequency of normal, mutated, and heterozygous alleles.
• Body Size, Coat Characteristics, and Sex Chromosomes: Details traits related to color, texture, size, and haplotypes for mitochondrial and Y chromosomes.
• Other Traits: Lists additional genetic traits specific to certain breeds, such as brachycephaly.

​
2. Genetics of Individuals (Module B)
Overview: Provides data for individual dogs on genetic diversity, relatedness, genetic “value,” and traits.

Features:

• Genetic Diversity: Displays inbreeding, fixation index, mean kinship, and heterozygosity for each dog.
• Traits & Disorders: Lists genes associated with traits and health issues, with options to filter and sort by genotypes.
• Genetic Ranks: Ranks dogs based on genetic diversity and relatedness.
• Genetic Relationships: Includes a dendrogram and heat map for visualizing genetic similarities among dogs.
• Kinship Matrix: Shows kinship levels with a color-coded heat map for assessing relatedness.
• Runs of Homozygosity (ROH): Highlights blocks of homozygosity to identify recent or historical inbreeding.
• Disease Risk Analysis: Uses a dendrogram to identify breed lines prone to specific diseases.
• Principal Components Analysis (PCA): Visualizes genetic subpopulations, helping identify genetic differences within a breed.

3. Test Mating (Module C)
Coming Soon: This feature will predict the level of inbreeding for potential litters from specific parent pairs.
​
Overall, the ICB Breeder Tool is a valuable resource for breeders seeking to make genetically informed decisions, minimizing genetic disorders while fostering breed diversity. It combines scientific rigor with practical tools, enabling breeders to track and analyze genetic health trends 

Glossary of Key Terms

• Allele: One of two or more alternative forms of a gene that arise by mutation and are found at the same place on a chromosome.
• Brachycephaly: A condition characterized by a shortened skull, often resulting in a flat-faced appearance.
• Dendrogram: A tree diagram that represents taxonomic or evolutionary relationships.
• Fixation Index: A measure of genetic differentiation among populations, often used to assess the level of inbreeding.
• Gene: A unit of heredity that is transferred from a parent to offspring and determines some characteristic of the offspring.
• Genotype: The genetic makeup of an individual organism.
• Haplotype: A set of DNA variations, or polymorphisms, that tend to be inherited together.
• Heterozygosity: The presence of two different alleles at a particular gene locus.
• Homozygosity: The presence of two identical alleles at a particular gene locus.
• Inbreeding: The mating of closely related individuals, leading to an increased chance of offspring inheriting harmful recessive traits.
• Kinship: A measure of the degree of genetic relatedness between two individuals.
• Mean Kinship: The average kinship coefficient between an individual and all other individuals in a population.
• Mitochondrial DNA: DNA located in the mitochondria, which is inherited maternally.
• Mutation: A change in the DNA sequence of a gene.
• Phenotype: The observable characteristics of an individual organism, resulting from the interaction of its genotype with the environment.
• Polymorphism: The presence of genetic variation within a population.
• Principal Components Analysis (PCA): A statistical method used to reduce the dimensionality of data by identifying principal components, which are linear combinations of the original variables.
• Runs of Homozygosity (ROH): Continuous stretches of homozygous genotypes within an individual's genome, indicating potential inbreeding.
• Y Chromosome: The sex chromosome that determines maleness in mammals.

Essential Concepts
These essential concepts are:

1) Inbreeding: probability of inheriting two copies of the same allele from an ancestor, which is called “homozygous”; it is also the fraction of genes that are homozygous. This measure of inbreeding is represented by the symbol F and is expressed either as a number between 0 and 1 (like 0.12), or as a percentage (like 12%).

2) Another way to express inbreeding is relative to the population. This is also called the “fixation index”, which is abbreviated as Fis. In a randomly breeding population, the average Fis is zero. A breeding of two individuals that are more closely related than average is considered inbreeding and will produce a positive Fis. Breeding two individuals that are less related than average is outbreeding, and Fis will be negative.

3) Heterozygosity is the fraction of genes for which the two alleles are different, so it is the opposite of homozygosity, in which the alleles are the same. Heterozygosity is represented by Ho.

4) Your relatives are your “kin”. In population genetics, we express the degree of relatedness between two individuals using the “kinship coefficient”, which is represented as the letter K.
​

Okay, armed with a few key bits of information, we’re ready to have a look at the components of the ICB Breeder Tool.

The Breeder Tool consists of three modules.

1) The first module provides information about the genetic status of the breed.

​2) The second module provides information about the genetics of individual dogs.

3) The third module provides predictions of the inbreeding of a litter produced by mating two particular individuals.

The first module Indicates that you are in the Breed Summary module in the upper left corner, and the breed is shown on the right.

This module provides a summary of the genetic information about the breed, specifically the genetic diversity, genetic disorders, and genetic traits. These are organized under a set of tabs.

Dog Inventory
​
​The first tab is labeled “Dogs”. It gives a quick inventory of the dogs currently in the database. It reports the total number of dogs in the data, as well as how many of those dogs are anonymous, which means that their identities are unknown, perhaps because they were part of a research study. Often, we only have basic DNA information for these data and data about genes for specific traits is not available. DNA contributed by the owner of a dog are “known” dogs, and for these we often have more comprehensive DNA information.

In this example, we have a database of 51 English Springer Spaniels. All of these are anonymous dogs and there are no dogs whose identity is known.

Genetic Diversity

The second tab provides a summary of information about genetic diversity in the breed based on the dogs in the database.

The measures of genetic diversity we use are the four we have described earlier: two types of inbreeding coefficient (F and Fis), kinship, and heterozygosity.

For each of these, the table reports the mean (or average), the median, and the maximum and minimum values.
​

Genotype Frequencies

​The next tab provides information about genes associated with health and diseases.

If all of the dogs in the dataset for a breed are anonymous, there might not be data for some of the rest of the tabs in this module, and the pages will be blank (or contain place-holders).

In this particular breed, dogs are tested for the genes for alanine aminotransferase and dilated cardiomyopathy. For both of these, there are two possible alleles. The chart reports the number of dogs in the sample that have two copies of the normal alleles (they are homozygous for the normal allele), and these are labeled “clear”. The chart also reports how many dogs have two copies of the alternative allele or mutation, indicated as homozygous for that allele, and the number of dogs that are heterozygous, with one copies of both the normal and the alternative allele (heterozygous).
​

Body Size

The next tab provides information about the genes associated with coat characteristics such as color, texture, or length. In this example, the K locus and E locus are genes for color.
​

Sex chromosomes

The next tab reports the information about the haplotypes for mitochondrial DNA and the Y chromosome that are found in that breed. There might be many of each of these in a breed, or just a few.​

​

Other Traits

​In some breeds there are genes for specific traits that are not found in most breeds, and these are under the “Other” tab. In this case, this breed carries genes associated with a shortened muzzle, a trait called brachycephaly.

For anonymous dogs, there will usually only be information under the first two tabs, for the inventory of the dogs and the statistics summarizing genetic diversity.

Module B provides the information for individual dogs about genetic diversity, relatedness to other dogs in the sample, the genetic "value" of a dog, the genes for traits and known mutations, and the genetic structure of the population. As in Module A, this information is organized on tabbed pages for navigation.
​

About the Tables
Many of the tables in the Breeder Tool have filtering and sorting features. The icon of the funnel in the ICB Code column indicates that you can filter this column to just the specific animals you want to see. The other columns can sorted by ascending or descending values by clicking on the little arrow next to the column label.
​

Genetic Diversity
Under the first tab, labeled “Genetic Diversity”, there is a table that summarizes the data for inbreeding, fixation index (Fis), the mean kinship, and heterozygosity of each dog. Each dog is given an ICB Code that corresponds to a key with the identity of each dog. These are the data that were used to create the "Genetic Diversity" graphs for the breed population in Module A.
​

Traits & Disorders
Under the next tab is a table containing the genetic information for known genes for traits and disorders. Depending on how many genes were tested, this can be a very wide table and you will need to scroll to the side to see all of it.

You will find it very useful to filter or sort by specific individuals or particular genotypes that you are interested in.

Genetic Ranks
​The Genetic Ranks tab presents charts that rank individual dogs by the value of a particular measure of genetic diversity or relatedness. Each bar is labeled with the ID of each dog. 

Where there are many individuals in the database, the ID codes of the dogs can be difficult or impossible to read. In some cases they are readable on a tablet where you can zoom in; in other cases, you can look up the value for a specific dog and use the scale on the y-axis to see where that dog would fall on the chart.

Genetic Relatiionships (heat map)
The next tab displays a colorful map of the genetic relationships among the dogs in the database in two forms: as a family tree called a dendrogram, where the dogs that are most similar genetically are clustered together, and as a matrix called a “heat map” that compares the genetic similarity of every pair of dogs in the database. The degree of relatedness is indicated by branch length in the dendrogram, and by color in the heat map.
​

Kinship Matrix
​Kinship, which is the degree of relatedness of two individuals, can also be displayed in this other type of heat map called a kinship matrix. This is similar to the heat map with the dendrogram, but in this one the dogs are not ordered. The advantage of this heat map is that the colors indicating the degree of kinship can be customized. In this one, a kinship coefficient of

6.25% or less is represented by green (equal to a cross of first cousins),

12.5% is yellow (half-sib cross), and 25% and above is red (full-sib cross). This allows you to quickly survey the levels of relatedness among animals in a population and also to identify specific levels that you might be interested in.
​

Runs of Homozygosity
The next tab, labeled “ROH”, contains charts that display “runs of homozygosity” on the chromosome. These are blocks of inbreeding that are the result of inbreeding. The genome is represented by arranging the chromosomes end to end from 1 to 38 across the top, and each row is the information for a particular dog. Blue represents blocks of homozygosity of a specific minimum length. Because older inbreeding tends to get broken up into smaller blocks, we can use the ROH charts to detect recent vs older inbreeding.
​

Disease Risk
​We can use a dendrogram that depicts the genetic relationships among individual dogs to explore the distribution of a particular trait or gene in the population.

For example, a research study was unable to identify a specific gene associated with mast cell tumors in Labrador retrievers. However, if we identify the dogs affected by mast cell tumors on the dendrogram, we might be able to identify lines that are predisposed to mast cell tumors. 

This technique might be useful for health issues as well as specific traits of interest.

Principal Components Analysis
​The last tab is for a graphical depiction of the genetic relationships among the dogs in the dataset using a statistical technique called “principal components analysis”, or PCA. This can be useful for identifying subpopulations of the breed that are genetically different such as show versus field lines of a breed, or varieties that differ by color.

Test Mating

Coming soon...Under construction

​The Test Mating page provides information for “predicted litter COI”, the average level of inbreeding expected in a litter produced by a specific pair of parents based on their kinship coefficient.