Merge pull request #573 from LStrachan/devel

Simplify vignette A

Author: gregorgorjanc

vignettes/A_Honeybee_biology.Rmd

@@ -50,18 +50,18 @@ library(package = "SIMplyBee")
 Figure 1 visualizes the initiation of the simulation. First, we simulate some
 honeybee genomes that represent the founder population. You can quickly generate
 random genomes using AlphaSimR's `quickHaplo()`. These founder genomes are
-rapidly simulated by sampling 0s and 1s, and do not include any species-specific
-demographic history. This is equivalent to all loci having allele frequency 0.5
-and being in linkage equilibrium. We use this approach only for demonstrations
-and testing.
+rapidly simulated by sampling chromosomes as series of 0s and 1s, and do not
+include any species-specific demographic history. This is equivalent to all loci
+having allele frequency 0.5 and being in linkage equilibrium. We use this approach
+only for demonstrations and testing.
 
 Alternatively, you can more accurately simulate honeybee genomes with
 SIMplyBee's `simulateHoneyBeeGenomes()`. This function simulates the honeybee
-genome using MaCS (Chen et al., 2009) for three subspecies: *A. m. ligustica*,
-*A. m. carnica*, and *A. m. mellifera* according to the demographic model
-described by Wallberg et al. (2014).
+genome using coalescent simulation of whole chromosomes using MaCS (Chen et al., 2009)
+for three subspecies: *A. m. ligustica*, *A. m. carnica*, and *A. m. mellifera*
+according to the demographic model described by Wallberg et al. (2014).
 
-As a first demonstration, we will use `quickHaplo()` and simulate genomes of two
+As a demonstration, we will use `quickHaplo()` and simulate genomes of two
 founding individuals. In this example, the genomes will be represented by only
 three chromosomes and 1,000 segregating sites per chromosome. Honeybees have 16
 chromosomes and far more segregating sites per chromosome, but we want a quick
@@ -71,43 +71,12 @@ simulation here.
 founderGenomes <- quickHaplo(nInd = 2, nChr = 3, segSites = 100)
 ```
 
-Alternatively, we use `simulateHoneyBeeGenomes()` to sample genomes of a founder
-population including 4 *A. m. mellifera* (North) individuals and 2 *A. m.
-carnica* individuals. The genomes will be represented by only three chromosomes
-and 5 segregating sites per chromosome. These numbers are of course extremely
-low because we want a quick examample for demonstrative reasons. This chunk of
-code should take a few minutes to run.
-
-```{r simulate honeybee genomes}
-founderGenomes2 <- simulateHoneyBeeGenomes(nMelN = 4,
-                                           nCar = 2,
-                                           nChr = 3,
-                                           nSegSites = 5)
-```
-
-Unfortunately, due to the complexity of this function, even using such small
-numbers takes a while to run. Simulating a group of founder genomes with more
-realistic numbers will therefore require a lot of time to run. We suggest
-running this part to an external server and save the outcome as an RData file,
-which we can load in our environment and work with it.
-
-```{r save Rdata file}
-# Save the genomes on a server
-save(founderGenomes2, file = "FounderGenomes2_3chr.RData")
-# Loade the saved genomes elsewhere
-load(file = "FounderGenomes2_3chr.RData")
-```
-
-Besides specifying the number of individuals, chromosomes, and segregating
-sites, `simulateHoneyBeeGenomes()`, also takes a number of genomic parameters:
-effective population size, ploidy, length of chromosomes in base pairs, genetic
-length of a chromosome in Morgans, mutation rate, recombination rate, and time
-of population splits. The default values for these numbers follow published
-references (Wallberg et al., 2014). While you can change these parameters, we
-don't advise doing this because such demographic models, and their parameters,
-are estimated jointly, so we should not be changing them independently. You can
-read more about these parameters in the help page
-`help(simulateHoneyBeeGenomes)`.
+As mentioned, the `simulateHoneyBeeGenomes()` generates more realistic chromosome
+samples, but also requires much more time. Hence, when you use `simulateHoneyBeeGenomes()`,
+we suggest you save the output to an RData file that you then load in your
+environment and work with it. See the function documentation using 
+`help(simulateHoneyBeeGenomes)` to learn all the parameters involved in the
+function.
 
 Now we are ready to setup global simulation parameters using `SimParamBee`.
 `SimParamBee` builds upon AlphaSimR's `SimParam`, which includes genome and