Introduction
Crossing the queen is the most complicated action in SIMplyBee, since mating of honeybees is versatile. Virgin queens can be mated in a controlled manner, either using artificial insemination or a mating station, or uncontrolled in an open mating setting. In the latter case, the mating depends on the distance between the virgin queens and the drones or drone producing colonies.
In this vignette, we first explain five ways of mating virgin queens and at the end we describe two special cases of mating virgin queens on a mating station and mating multiple virgin queens in different ways.
Crossing decision tree
To facilitate the understanding of the crossing function and choosing the correct option for your need, we created a diagram with a decision tree following the questions:
- How many virgin queens/colonies are you mating?
- one virgin queen (Pop-class with one individual) or one virgin colony (Colony-class or MultiColony-class with a single colony). This is marked asoption 1 on the diagram. For this, you need to provide the females as either
Pop,Colony,MultiColony class and males as a singlePop class object. - multiple virgin queens (Pop-class with multiple individuals) or multiple virgin colonies (MultiColony-class with multiple colonies). Continue to question 2.
- If you are matingmultiple virgin queens/colonies, do you want to:
- cross withdrone packages, that is pre-prepared populations of drones for each queen? You probably want to use this if you want more control over the drones that mate with each queen, such as in artificial insemination. This crossing option is suitable when mating a smaller number of virgin queens since it is computationally not the most efficient. For this option, you need to provide the females as either
Pop,Colony,MultiColony class and males as a list ofPop class object - one for each virgin queen/colony. - crossaccording to a cross plan. This option is the recommended option when mating a larger number of virgin queens/colonies. Continue to question
- When matingaccording to a cross plan, you can either:
- provide a cross plan, which is a list where the list names are either virgin queens’ or virgin colonies IDs and the content for each virgin queen/colony is either
- the IDs drones she will mate with; in this case, you also have to provide
drones argument as a singlePop class that represents the DCA with all the drones; - the IDs of the drone producing colonies that she will mate with; in this case, you also have to provide the
droneColonies argument as aMultiColony class with all drone producing colonies. You probably want to use this option when you are simulating an existing pedigree or trying to have more control over mating for a larger number of queens, such as on a mating stations. You can also use an auxiliary functioncreateCrossPlan() to create the cross plan, but in this case, you probably want to move to question 4.
- create the cross plan internally, either by sampling the drones or drone producing colonies randomly or spatially. Continue to question 4.
- When matingaccording to an internally created cross plan, you can either:
- cross according to arandom cross plan, where the mates for mating are sampled randomly. For this, you need to provide the females as either
Pop,Colony,MultiColony class and males either through thedrones argument as aPop class object, or through thedroneColonies argument as aMultiColony class object. You probably want to use this option if you are not concerned with spatial simulation or are simulation a small area with a single DCA. You can also use this option to mate virgin queens on a mating station in combination with thecreateMatingStationsDCA() function. - cross according to aspatial cross plan, where the mates for mating are sampled according to their geographic location in a set radius. For this, you need to provide both females and males as
MultiColony objects withthe location set, hence males have to be provided through thedroneColonies argument. You probably want to use this option when simulating realistic breeding programs and populations.
Start by loading the package:
library(package ="SIMplyBee")library(package ="ggplot2")
First, we create a founder population and some virgin queen and virgin colonies for crossing. We will also create fourMultiColony objects that will represent colonies from four different beekeepers.
# Simulate 40 founder genomesfounderGenomes <-quickHaplo(nInd =60,nChr =1,segSites =100)# Set global population paramatersSP <-SimParamBee$new(founderGenomes)# Create a base population of 40 virgin queensbasePop <-createVirginQueens(founderGenomes)# Prepare populations with a single virgin queenvirginQueen1 <-basePop[1]# Prepare populations with multiple virgin queensvirginQueens1 <-basePop[2:4]# Prepare virgin MultiColony objectsbeekeeper1 <-createMultiColony(basePop[6:11])beekeeper2 <-createMultiColony(basePop[12:16])beekeeper3 <-createMultiColony(basePop[17:21])beekeeper4 <-createMultiColony(basePop[22:31])beekeeper5 <-createMultiColony(basePop[32:40])beekeeper6 <-createMultiColony(basePop[41:50])beekeeper7 <-createMultiColony(basePop[51:60])
We will now create a groups of drones from the remaining queens with 1,000 drones per queen that will represent a drone congregation area (DCA).
# Create a DCA from the remaining virgin queensDCA <-createDrones(basePop[42:50],nInd =1000)
Option2: Cross with drone packages (pre-selecting drone populations)
We continue by crossing our virgin females to drone packages, that is a pre-selected populations of drones. We recommend this option when mating a larger number of queens, although it could be used for mating a single virgin queen/colony as well, and when you want more control over the drone selection. We pre-select the groups by pulling a desired number of drone packages from a DCA with the functionpullDroneGroupsFromDCA(). This function requires you to specify a group of drones (DCA), how many groups you want to pull from the DCA (n), and how many drones per group you want (nDrones). FornDrones, you can either specify an integer or a sampling function, which results in a different number of drones in each of the pulled groups (you can read more about this in the Sampling functions vignette). These sampling functions are particularly useful in crossing simulations:
nFathersPoisson(): samples the number of drones from a Poisson distribution with a default mean of 15 (the user can specify a different mean) - the output can contain the value 0 andnFathersTruncPoisson(): samples the number of drones from a zero truncated Poisson distribution with a default mean of 15 (the user can specif a different mean) - the output does not contain the value 0.
If these functions do not satisfy your needs, you can specify your own sampling function(s).
We can pull the drone groups out separately for each crossing or pull them out all at once.
# Pre-select drone (father) populations from a DCAdroneGroups <-pullDroneGroupsFromDCA(DCA,n =20,nDrones = nFathersTruncPoisson)
There is a different number of drones in each group because we used a sampling function:
sapply(droneGroups,FUN = nInd)#> [1] 17 17 21 13 17 13 15 17 15 13 17 17 23 10 13 14 10 16 11 12
Now, we can cross our virgin queens to drone groups.
# Pop-class with multiple queensqueens1 <-cross(x = virginQueens1,drones = droneGroups[1:3],checkCross ="warning")nFathers(queens1)#> [1] 17 17 21
# MultiColony-classbeekeeper1 <-cross(x = beekeeper1,drones = droneGroups[4:9],checkCross ="warning")nFathers(beekeeper1)#> 1 2 3 4 5 6#> 13 17 13 15 17 15
Option 3: Cross according to a provided cross plan
When mating a large number of queens, creating drone populations can be computationally demanding! A better option is to cross according to a cross plan that includes all the females and the males that they will mate with, and males as a single object. The cross plan is a list whose names are the IDs of either the virgin queens or the colonies you want to cross (but not both in the same cross plan!), and the content for each queen is either the IDs of drone or drone producing colonies she will mate with (again, use either only drone or only drone producing colony’s IDs in one cross plan!). When providing drone IDs, you also have to provide the males through thedrones argument as a singlePop class that represents the DCA with all the drones. The DCA can represent a DCA of randomly assembled drones in the vicinity, a ‘global’ DCA of all drones available at a time, or a DCA at a mating station. If you are providing drone producing colonies’ IDs, you need to provide the males through thedroneColonies argument as a singleMultiColony class. Mating with drone producing colonies might be computationally more efficient since only the drones needed for mating are created on the spot.
You can use the cross plan either to gain more control over mating and create the cross plan in a manual manner, or to simply reduce the computational burden when mating a large number of virgin queens. For that, you can either use the functioncreateCrossPlan(), if you want to inspect the cross plan before deployment, or simply let thecross() function to create the cross plan internally (see options 5 and 6). The latter is just a shortcut since it calls thecreateCrossPlan() function from within thecross(). Both option hence allow to create the cross plan by randomly sampling available drones or drone producing queens, or sample them according to their geographical proximity to the virgin colony. Do note that to perform spatial sampling, the females and males have to be provided andMultiColony objects with a set location. Another situation where you would want to use the cross plan is when you want.
NOTE: You can create one cross plan for a number of virgin queens or colonies and then mate each one separately. Let’s say you have 100 colonies from two beekeepers. You can create a cross plan for all of them in one go, and then mate colonies from each beekeeper separately by providing the same cross plan. Do note again, that one cross plan can either contain only virgin queen or only virgin colonies’ IDs, no mix and and match!
Let us now create a random cross plan with drone IDs to mate the virgin colonies from one of the beekeepers.
crossPlan1 <-createCrossPlan(x = beekeeper2,drones = DCA,nDrones = nFathersPoisson)# Inspect the cross plancrossPlan1#> $`7`#> [1] "7847" "4000" "2130" "481" "4054" "5267" "3811" "1980" "3833" "7920"#> [11] "6255" "7649" "8789"#>#> $`8`#> [1] "3625" "7296" "8221" "925" "8568" "4391" "384" "2269" "172" "8000"#> [11] "8326"#>#> $`9`#> [1] "952" "3256" "4384" "1224" "8026" "1271" "3703" "6850" "3759" "2031"#> [11] "1242"#>#> $`10`#> [1] "4942" "2734" "756" "7345" "7100" "5265" "6386" "2685" "5924" "2321"#> [11] "5022" "2456" "1804"#>#> $`11`#> [1] "4972" "4824" "6875" "4276" "3382" "185" "8399" "3815" "7280" "8514"#> [11] "3523" "5879"sapply(crossPlan1, length)#> 7 8 9 10 11#> 13 11 11 13 12
We see that the cross plan is a list, with names being colony IDs and elements of the list being the ID of drones that each virgin colony will mate with. Now, we can cross the colonies of each beekeeper by providing thecrossPlan1 to thecross() function and the created DCA through thedrones argument.
# Cross the colonies of the beekeeper 2beekeeper2 <-cross(x = beekeeper2,drones = DCA,crossPlan = crossPlan1,checkCross ="warning")nFathers(beekeeper2)#> 7 8 9 10 11#> 13 11 11 13 12
Now, let us now create a spatial cross plan, this time by sampling drone producing colonies that will provide drones to mate with another one of our virgin apiaries. We will mate colonies of beekeeper 3. In this examples, colonies from beekeeper 1 and beekeeper 2 will serve as drone producing colonies. Although we are not sampling drones directly, you still need to providenDrones argument to determine the number of sampled drone producing colonies.
For spatial mating, we need to set the location to our apiaries. Here, we are reading the locations in from a file with the coordinates of all colonies, then selecting the colonies of a specific beekeeper, creating a list of coordinate pairs (with theMap function). We could also set locations randomly.
# Read locations from a filelocations =read.csv("Colony_locations.csv")# Set location to apiariesbeekeeper1_locations <-locations[locations$ColonyID%in%getId(beekeeper1),]beekeeper1 <-setLocation(beekeeper1,location =Map(c, beekeeper1_locations$X, beekeeper1_locations$Y))beekeeper2_locations <-locations[locations$ColonyID%in%getId(beekeeper2),]beekeeper2 <-setLocation(beekeeper2,location =Map(c, beekeeper2_locations$X, beekeeper2_locations$Y))beekeeper3_locations <-locations[locations$ColonyID%in%getId(beekeeper3),]beekeeper3 <-setLocation(beekeeper3,location =Map(c, beekeeper3_locations$X, beekeeper3_locations$Y))
We can plot the location of colonies to see how they arrange in space. Again, we want to mate the virgin queens of beekeeper 3 with drone produced by colonies of beekeepers 1 and 2.
locationsDF <-data.frame(Location =getLocation(c(beekeeper1, beekeeper2, beekeeper3),collapse =TRUE),Beekeeper =c(rep("Beekeeper1",nColonies(beekeeper1)),rep("Beekeeper2",nColonies(beekeeper2)),rep("Beekeeper3",nColonies(beekeeper3))))ggplot(data = locationsDF,aes(x = Location.1,y = Location.2,colour = Beekeeper))+geom_point()
Next, we create a spatial cross plan for mating our colonies. We set the radius to however far we believe our drone can fly according to the scale of locations coordinates. All the drones within that area then have the same probability of mating with the queen, which follows the believe that drones aggregate in a DCA.
crossPlan2 <-createCrossPlan(x = beekeeper3,droneColonies =c(beekeeper1, beekeeper2),spatial =TRUE,radius =3,nDrones =13)# Inspect the cross plancrossPlan2#> $`12`#> [1] "10" "7" "5" "3"#>#> $`13`#> [1] "5" "4"#>#> $`14`#> [1] "5" "10" "4"#>#> $`15`#> [1] "9" "4" "3" "6" "2" "8"#>#> $`16`#> [1] "3" "1" "11" "10"sapply(crossPlan2, length)#> 12 13 14 15 16#> 4 2 3 6 4
We see that there is a different number of drone producing queen for each virgin queen in the specified radius . It could also happen, there are no drone producing colonies in that radius. In that case, you can either increase the radius, if you think that is sensible, or some virgin queens will remain unmated. Once we have the spatial cross plan, we can mate our colonies.
beekeeper3 <-cross(x = beekeeper3,crossPlan = crossPlan2,droneColonies =c(beekeeper1, beekeeper2),nDrones =13,checkCross ="warning")# Inspect the number of fathersnFathers(beekeeper3)#> 12 13 14 15 16#> 13 13 13 13 13
As you see, we have duplicated quite some information in running first a function that creates a cross plan and then performing the crossing. If you are not really keen on inspecting the cross plan before mating, we suggest you use the options 4 and 5 (see below).
You may also want to provide an existing cross plan if you want to create more tailored cross plans and cross virgin queens with different methods (i.e. you want to mate one queen in the apiary in an artificial insemination, some of them would be open mated etc.). For that, see the section “Special cases” at the end of this vignette.
Options 4 and 5: Cross according to an internally created cross plan
The procedure behind this option is exactly the same as described in option 4 with usingcreateCrossPlan() function to create either a random and spatial cross plan and then crossing according to the plan. The difference is that in this option, the two steps are combined within thecross() function.
Let us mate the colonies of the remaining beekeepers 4 and 5 with colonies of beekeepers 1 to 3 as drone producing colonies. Let’s first set the locations of the beekeepers 4 and 5 colonies.
beekeeper4_locations <-locations[locations$ColonyID%in%getId(beekeeper4),]beekeeper4 <-setLocation(beekeeper4,location =Map(c, beekeeper4_locations$X, beekeeper4_locations$Y))beekeeper5_locations <-locations[locations$ColonyID%in%getId(beekeeper5),]beekeeper5 <-setLocation(beekeeper5,location =Map(c, beekeeper5_locations$X, beekeeper5_locations$Y))
Now, we will mate the colonies of beekeeper 4 in a random manner and colonies of beekeeper 5 in a spatial manner, all within thecross() function. For this, you have to pass a keyword “create” to thecrossPlan argument in thecross().
beekeeper4 <-cross(x = beekeeper4,droneColonies =c(beekeeper1, beekeeper2, beekeeper3),crossPlan ="create",spatial =FALSE,nDrones =12,checkCross ="warning")#> [1] "Cross plan, drone colonies"nFathers(beekeeper4)#> 17 18 19 20 21 22 23 24 25 26#> 12 12 12 12 12 12 12 12 12 12beekeeper5 <-cross(x = beekeeper5,droneColonies =c(beekeeper1, beekeeper2, beekeeper3),crossPlan ="create",spatial =TRUE,radius =3,nDrones =12,checkCross ="warning")nFathers(beekeeper5)#> 27 28 29 30 31 32 33 34 35#> 12 12 12 12 12 12 12 12 12