Introduction

This vignette demonstrates how to use SIMpoly to simulate multiple biparental populations and how you can use mappoly2 to construct a consensus genetic map. The workflow includes:

1. Defining simulation parameters and pedigree.
2. Simulating multiple chromosomes using SIMpoly.
3. Constructing linkage maps using mappoly2.
4. Integrating maps to generate a consensus map.

Note: The consensus map generation requires mappoly2. If you only have SIMpoly, the first simulation steps will work, but the final consensus map will not be generated.

---

Step 1: Installing and Loading Packages

library(SIMpoly)
library(mappoly2)  # for consensus mapping

---

Step 2: Define Simulation Parameters and Pedigree

# Define ploidy for each parent.
ploidy.vec <- c(4, 2, 4, 2, 4, 4)  # for six parents
names(ploidy.vec) <- c("P1", "P2", "P3", "P4", "P5", "P6")

# Define parental crosses.
parents.mat <- matrix(c("P1", "P2",
                        "P1", "P3",
                        "P2", "P2",
                        "P3", "P4",
                        "P4", "P5",
                        "P5", "P6",
                        "P5", "P2"),
                      ncol = 2, byrow = TRUE)

# Define number of offspring per cross.
n_offspring <- c(200, 100, 200, 100, 300, 200, 100)

# Define allele set for each founder.
alleles <- list(P1 = 0:1, P2 = 0:1, P3 = 0:1, P4 = 0:1, P5 = 0:1, P6 = 0:1)

# Generate pedigree.
pedigree <- simulate_pedigree(ploidy.vec, parents.mat, n_offspring)
head(pedigree, n = 10)
#>      parent_1 parent_2 individual cross
#> 1        <NA>     <NA>         P1 NAxNA
#> 2        <NA>     <NA>         P2 NAxNA
#> 3        <NA>     <NA>         P3 NAxNA
#> 4        <NA>     <NA>         P4 NAxNA
#> 5        <NA>     <NA>         P5 NAxNA
#> 6        <NA>     <NA>         P6 NAxNA
#> P1.1       P1       P2    P1xP2_1 P1xP2
#> P1.2       P1       P2    P1xP2_2 P1xP2
#> P1.3       P1       P2    P1xP2_3 P1xP2
#> P1.4       P1       P2    P1xP2_4 P1xP2

---

Step 3: Simulate Multiparental Data

n.chr <- 3  # Number of chromosomes
map.len <- c(100, 120, 90)  # Chromosome lengths
n_mrk <- c(2000, 1500, 2000)  # Number of markers per chromosome

# Simulate genetic data
result <- simulate_multiparental_data(n.chr, map.len, pedigree, ploidy.vec, n_mrk, alleles,  missing = 0.1, p = .3 , rho = .7, seed = 1)

# Extract results
wide_df <- result$wide_df
all_dat <- result$dat
parent_homologs <- result$parent_homologs[[1]]
plot_parent_phase_gg(parent_homologs)

plot_correlated_sets(result$counts)

---

Step 4: Construct Linkage Maps with mappoly2

MAPs <- vector("list", nrow(parents.mat))

for(i in 1:nrow(parents.mat)){
  p1 <- parents.mat[i,1]
  p2 <- parents.mat[i,2]
  dat <- mappoly2:::table_to_mappoly(all_dat[[i]], ploidy.vec[p1], ploidy.vec[p2], p1, p2)
  
  # Filtering steps
  dat <- filter_markers_by_chisq_pval(dat, plot = FALSE)
  dat <- filter_markers_by_missing_rate(dat, mrk.thresh = .1, plot = FALSE)
  dat <- filter_individuals_by_missing_rate(dat, ind.thresh = .1, plot = FALSE)
  dat <- pairwise_rf(dat, mrk.scope = "all", ncpus = 8)
  plot(dat)
  dat <- rf_filter(dat, probs = c(0, 1), diagnostic.plot = FALSE)
  
  # Group and order markers
  g <- group(dat, expected.groups = n.chr, comp.mat = TRUE, inter = FALSE)
  plot(g)
  s <- make_sequence(g, ch = list(1,2,3))
  s <- order_sequence(s, type = "genome")
  s <- pairwise_phasing(s, type = "genome", parent = "p1p2")
  s <- mapping(s, type = "genome", parent = "p1p2", ncpus = n.chr)
  s <- calc_haplotypes(s, type = "genome", ncpus = n.chr)
  
  MAPs[[i]] <- s
}
plot_map(MAPs[[1]], lg = 1, type = "genome")

---

Step 5: Construct a Consensus Map

# Visualize multiple maps
plot_multi_map(MAPs)

# Integrate maps
w <- prepare_to_integrate(MAPs, type = "genome")
plot(w)

# Estimate consensus map
x <- estimate_consensus_map(w, ncpus = 8)
plot(x)
plot(x, only.consensus = TRUE, col = mappoly::mp_pallet3(3))

# Compute haplotypes
system.time(x <- calc_consensus_haplo(x, ncpus = 8))

# Visualize consensus haplotypes
plot_consensus_haplo(x, lg = 1, ind = "P1xP2_1")
plot_consensus_haplo(x, lg = 1, ind = "P5xP6_1")

---

Summary

This vignette demonstrated how to: 1. Define and simulate multiparental populations using SIMpoly. 2. Construct linkage maps using mappoly2. 3. Generate and visualize a consensus map.

SIMpoly - Simulation of Genetic Marker Data in Pedigreed Populations

Marcelo Mollinari

🚀 Introduction

This vignette demonstrates how to use SIMpoly to simulate multiple biparental populations and how to leverage mappoly2 to construct a consensus genetic map. The workflow includes:

1. Installing from GitHub
2. Defining simulation parameters and pedigree.
3. Simulating multiple chromosomes using SIMpoly.
4. Constructing linkage maps using mappoly2.
5. Integrating maps to generate a consensus map.

⚠️ Note: The consensus map generation requires mappoly2. If you only have SIMpoly, the first simulation steps will work, but the final consensus map will not be generated.

📦 Installation from GitHub

You can install the development version from GitHub. Within R, install devtools:

install.packages("devtools")

If you are using Windows, please install the latest recommended version of Rtools.

To install SIMpoly from GitHub, use:

devtools::install_github("mmollina/SIMpoly", dependencies=TRUE)

---

🛠️ Step 1: Installing and Loading Packages

library(SIMpoly)
library(mappoly2)  # for consensus mapping

---

🔬 Step 2: Define Simulation Parameters and Pedigree

# Define ploidy for each parent.
ploidy.vec <- c(4, 2, 4, 2, 4, 4)  # for six parents
names(ploidy.vec) <- c("P1", "P2", "P3", "P4", "P5", "P6")

# Define parental crosses.
parents.mat <- matrix(c("P1", "P2",
                        "P1", "P3",
                        "P2", "P2",
                        "P3", "P4",
                        "P4", "P5",
                        "P5", "P6",
                        "P5", "P2"),
                      ncol = 2, byrow = TRUE)

# Define number of offspring per cross.
n_offspring <- c(200, 100, 200, 100, 300, 200, 100)

# Define allele set for each founder.
alleles <- list(P1 = 0:1, P2 = 0:1, P3 = 0:1, P4 = 0:1, P5 = 0:1, P6 = 0:1)

# Generate pedigree.
pedigree <- simulate_pedigree(ploidy.vec, parents.mat, n_offspring)
head(pedigree, n = 10)

---

📊 Step 3: Simulate Multiparental Data

n.chr <- 3  # Number of chromosomes
map.len <- c(100, 120, 90)  # Chromosome lengths
n_mrk <- c(2000, 1500, 2000)  # Number of markers per chromosome

# Simulate genetic data
result <- simulate_multiparental_data(n.chr, map.len, pedigree, ploidy.vec, n_mrk, alleles, missing = 0.1, p = .3, rho = .7)

# Extract results
wide_df <- result$wide_df
all_dat <- result$dat
parent_homologs <- result$parent_homologs[[1]]
plot_parent_phase_gg(parent_homologs)

---

🔗 Step 4: Construct Linkage Maps with mappoly2

MAPs <- vector("list", nrow(parents.mat))

for(i in 1:nrow(parents.mat)){
  p1 <- parents.mat[i,1]
  p2 <- parents.mat[i,2]
  dat <- mappoly2:::table_to_mappoly(all_dat[[i]], ploidy.vec[p1], ploidy.vec[p2], p1, p2)
  
  # Filtering steps
  dat <- filter_markers_by_chisq_pval(dat, plot = FALSE)
  dat <- filter_markers_by_missing_rate(dat, mrk.thresh = .1, plot = FALSE)
  dat <- filter_individuals_by_missing_rate(dat, ind.thresh = .1, plot = FALSE)
  dat <- pairwise_rf(dat, mrk.scope = "all", ncpus = 8)
  dat <- rf_filter(dat, probs = c(0, 1), diagnostic.plot = FALSE)
  
  # Group and order markers
  g <- group(dat, expected.groups = n.chr, comp.mat = TRUE, inter = FALSE)
  s <- make_sequence(g, ch = list(1,2,3))
  s <- order_sequence(s, type = "genome")
  s <- mapping(s, type = "genome", parent = "p1p2", ncpus = n.chr)
  
  MAPs[[i]] <- s
}
plot_map(MAPs[[1]], lg = 1, type = "genome")

---

🏆 Acknowledgment

This package has been developed as part of the project AFRI-Grant: A Genetics-Based Data Analysis System for Breeders in Polyploid Breeding Programs and SCRI-Grant: Tools for polyploids, funded by USDA NIFA.