Introduction to the Data Set

---

Figure 1

Image 1 of 1: ‘Figure showing Type 2 diabetes & insulin.’

Created in BioRender.com

---

Figure 2

Image 1 of 1: ‘Figure showing intercross breeding design.’

---

Input File Format

---

Figure 1

Image 1 of 1: ‘Table showing the mouse genotypes as BB, BR, and RR.’

Attie Sample Genotypes

---

Figure 2

Image 1 of 1: ‘Table showing the marker, chromosome, and centimorgan position for five markers’

Attie Genetic Map

---

Figure 3

Image 1 of 1: ‘Table showing top five rows of physical marker map.’

Attie Physical Map

---

Figure 4

Image 1 of 1: ‘Table showing top five rows of phenotype table, including insulin’

Attie Phenotypes

---

Figure 5

Image 1 of 1: ‘Table showing the top five rows of covariates table’

Attie Covariates

---

Figure 6

Image 1 of 1: ‘Figure showing the qtl2 control file’

Attie Control File

---

Calculating Genotype Probabilities

---

Figure 1

Image 1 of 1: ‘a chromosome with two typed markers labeled BR and RR with a locus of unknown genotype between them’

Genotype probabilities must be calculated between typed markers; adapted from Broman & Sen, 2009

---

Figure 2

Image 1 of 1: ‘Figure showing three-dimensional array of genotype probabilities (genoprobs)’

Three-dimensional genotype probabilities array

---

Figure 3

Image 1 of 1: ‘a web page showing R data structures including one-dimensional vectors and lists, two dimensional dataframes and matrices, and n-dimensional arrays’

Notice that arrays in R require data to be all of the same type - all numeric, all character, all Boolean, etc. If you are familiar with data frames in R you know that you can mix different kinds of data in that structure. The first column might contain numeric data, the second column character data, the third Boolean (True / False), and so on. Arrays won’t accept mixed data types though.

---

Figure 4

---

Figure 5

Image 1 of 1: ‘a workflow diagram showing genotypes and genetic maps used to calculate genotype probabilities’

We have completed the first steps of the mapping workflow.

---

Performing a Genome Scan

---

Figure 1

Image 1 of 1: ‘Null and alternative hypotheses’

Linear regression can identify QTL; adapted from Broman & Sen, 2009

---

Figure 2

Image 1 of 2: ‘Alternative hypothesis’

Image 2 of 2: ‘Null hypothesis’

---

Figure 3

Image 1 of 1: ‘Haley Knott regression’

Haley-Knott regression for missing genotypes

---

Figure 4

Image 1 of 1: ‘Figure showing sex and additive QTL effects.’

Additive Sex Effect

---

Figure 5

Image 1 of 1: ‘Figure showing sex and additive QTL effects.’

Additive Sex Effect

---

Figure 6

---

Figure 7

Image 1 of 1: ‘A workflow diagram showing genotypes and genetic map used to calculate genotype probabilities which are then used to perform a genome scan along with phenotypes, covariates, and an optional physical map’

We have completed these steps in the mapping workflow.

---

Figure 8

---

Figure 9

Image 1 of 1: ‘Figure showing interactive sex by QTL effects’

Interactive Sex by QTL Effect

---

Figure 10

---

Figure 11

---

Figure 12

---

Calculating A Kinship Matrix

---

Figure 1

---

Figure 2

Image 1 of 1: ‘We have completed these steps in the mapping workflow.’

{alt=“Part of the mapping workflow starting with genotypes and a genetic map used to calculate genotype probabilities, which are then used to calculate kinship between individuals in the data.} By default, the genotype probabilities are converted to allele probabilities, and the kinship matrix is calculated as the proportion of shared alleles. Also by default we omit the X chromosome and only use the autosomes. To include the X chromosome, use omit_x=FALSE.

---

Figure 3

---

Performing a genome scan with a linear mixed model

---

Figure 1

Image 1 of 1: ‘A world map showing the origins and ranges of five Mus musculus subspecies: domesticus, musculus, castaneus, bactrianus and molossinus’

Geographical ranges of subspecies in the Mus musculus group. From Mouse Genetics: Concepts and Applications by Lee M. Silver, Oxford University Press, 1995.

---

Figure 2

Image 1 of 1: ‘A genealogy chart showing the origin of commonly used inbred strains as Chinese, Japanese and English mouse fanciers. Abby Lathrop of Granby Mouse Farm supplied fancy mice to researchers including Marsh, C.C. Little, MacDowell, Castle and Strong in the early 1900s. These mice were developed into common inbred strains such as C57BL, 129, DBA and C3H strains.’

The genealogy of inbred mice. From Origins of Inbred Mice edited by Herbert C. Morse III, Academic Press, 1978. Permission from Elsevier.

---

Figure 3

Image 1 of 1: ‘Null versus alternative hypotheses’

Null and Alternative Hypotheses

---

Figure 4

---

Figure 5

---

Figure 6

---

Performing a Genome Scan with Binary Traits

---

Figure 1

Image 1 of 1: ‘C57BL/6J mouse’

C57BL/6J

---

Figure 2

Image 1 of 1: ‘BTBR’

BTBR

---

Figure 3

Image 1 of 1: ‘C3H/HeJ mouse’

C3H/HeJ

---

Figure 4

Image 1 of 1: ‘Image showing tufted hair loss in BTBR mouse’

Tufted hair loss pattern

---

Figure 5

---

Figure 6

---

Figure 7

---

Finding Significant Peaks via Permutation

---

Figure 1

---

Figure 2

Image 1 of 1: ‘Figure showing decreasing variance of significance threshold estimates with increasing permutations’

Significance Threshold Variance

---

Figure 3

Image 1 of 1: ‘A diagram showing mapping steps including calculating genotype probabilities, calculating kinship, performing a genome scan, finding QTL peaks, and performing a permutation test.’

As with scan1(), you can speed up the calculations on a multi-core machine by specifying the argument cores. With cores=0, the number of available cores will be detected via parallel::detectCores(). Otherwise, specify the number of cores as a positive integer. For large data sets, be mindful of the amount of memory that will be needed; you may need to use fewer than the maximum number of cores, to avoid going beyond the available memory.

---

Figure 4

Image 1 of 1: ‘Comparison of significance thresholds’

Significance Thresholds with/without Kinship

---

Finding QTL peaks

---

Figure 1

---

Estimating QTL effects

---

Figure 1

Image 1 of 1: ‘Null and alternative hypotheses’

---

Figure 2

---

Figure 3

---

Figure 4

---

Figure 5

---

Figure 6

---

Figure 7

Image 1 of 1: ‘A diagram showing mapping steps up to estimating QTL effects.’

::::::::::::::::::::::::::::::::::::: challenge

---

Figure 8

---

Figure 9

---

Figure 10

Image 1 of 1: ‘Figure showing additive & dominance QTL effects’

Additive & Dominance Effects

---

Figure 11

---

Figure 12

---

Figure 13

---

Integrating Gene Expression Data

---

Figure 1

---

Figure 2

---

Figure 3

---

Figure 4

---

Figure 5

---

QTL Mapping in Diversity Outbred Mice

---

Figure 1

Image 1 of 1: ‘Figure showing the colors and letter codes of the CC/DO founders.’

CC and DO Founder Alleles

---

Figure 2

Image 1 of 1: ‘Figure showing CC and DO genomes’

Example Collaborative Cross and Diversity Outbred genomes

---

Figure 3

Image 1 of 1: ‘Benzene study dosing showing 6 hours per day, 5 days per week of inhalation.’

Benzene Study Dosing

---

Figure 4

---

Figure 5

---

Figure 6

---

Figure 7

Image 1 of 1: ‘Three-dimensional figure of genoprobs list.’

Genoprobs Object

---

Figure 8

---

Figure 9

Image 1 of 1: ‘a table showing the probabilities for each of 36 genotypes in the Diversity Outbred followed by a second table showing probabilities for each of the 8 founder alleles’

Genotype and allele probabilities

---

Figure 10

---

Figure 11

---

Figure 12

---

Figure 13

---

Figure 14

---

Figure 15

---

Figure 16

---

Figure 17

Image 1 of 1: ‘Figure showing haplotypes being used to impute founder SNPs onto DO genomes’

SNP Imputation

---

Figure 18

Image 1 of 1: ‘Plot showing LOD of each SNP in QTL interval’

This plot shows the LOD score for each SNP in the QTL interval. The SNPs occur in “shelves” because all of the SNPs in a haplotype block have the same founder strain pattern. The SNPs with the highest LOD scores are the ones for which CAST/EiJ contributes the alternate allele.

---

Figure 19

Image 1 of 1: ‘Plot showing LOD of each SNP in QTL interval with genes below’

Now that we have the genes in this interval, we would like to know which founders have the minor allele for the SNPs with the highest LOD scores. To do this, we will highlight SNPs that are within a 1 LOD drop of the highest LOD and we will add an argument to show which founder contributes the minor allele at the highlighted SNPs.

---

Figure 20

Image 1 of 1: ‘Plot showing LOD of each SNP in QTL interval with genes below and strain distribution pattern above’

Bone Marrow MN-RET Association Mapping

---

Figure 21

Image 1 of 1: ‘Figure showing LOD plots of Sult3a1 & 2 and CAST-specific haplotype effects’

The plot above shows the genome scane for two genes: Sult3a1 and Gm4794. Gm4794 has been renamed to Sult3a2. As you can see, both Sult3a1 Sult3a2 have eQTL in the same location at the MN-RET QTL on chromosome 10. Mice carrying the CAST allele (in green) express these genes more highly. Sult3a1 is a sulfotransferase that may be involved in adding a sulfate group to phenol, one of the metabolites of benzene.

---

Figure 22

Image 1 of 1: ‘Figure showing that Sult3a1 & 2 are paralogs’

Ensembl gene tree of Sult3a1 & Sult3a2

---

Figure 23

Image 1 of 1: ‘Ensembl viewer showing genes and structural variants under that chromosome 10 QTL peak’

Ensembl Structural Variants

---

Figure 24

Image 1 of 1: ‘Ensembl viewer showing genes and structural variants under that chromosome 10 QTL peak’

Ensembl Structural Variants

---

Figure 25

Image 1 of 1: ‘Figure showing CAST with a duplication at the Sult3a1/2 locus.’

DO Founder BAM file pileups

---

Figure 26

Image 1 of 1: ‘Figure showing benezene metabolism pathways’

Benezene metabolism pathways

---

Figure 27

Image 1 of 1: ‘Figure showing mice carrying the CAST allele being protected from benezene toxicity’

Benzene metabolism hypothesis

---

Figure 28

---

Figure 29

---

Figure 30

---

Figure 31

Image 1 of 1: ‘Figure showing pre-dose MN-RET association mapping with WSB carrying the alternate allele’

Pre-dose MN-RET Association Mapping

---