sim_microbiota: Simulate microbial alignments according to different models...

In BPerezLamarque/HOME: Phylogenetic Models of HOst-Microbiota Evolution

Simulate microbial alignments according to different models of Host-Microbiota evolution

Description

This function simulates microbial alignments according to different models of host-microbiota evolution (strict vertical transmission, horizontal transmission, environmental acquisition).

Usage

sim_microbiota(name, name_index, simul, mu=1, n=20, provided_tree=NULL, N=300,
proportion_variant=0.1, model="uniform", mean=0.5, sd=0.01, host_signal=10,
geo_signal=0, stochastic_map=NULL, delta=0,
path=getwd(), seed=1, nb_cores=1, ...)

Arguments

name	the name of the run
name_index	a vector with the names of the different symbionts (e.g. name of the OTUs) to simulate
simul	a vector of scenarii for each simulated alignment ("0" for strict vertical transmission, any positive integer for the number of host-switches transmissions, and "indep" to simulate independent evolutions and environmental acquisitions). This vector must have the same length as "name_index"
provided_tree	optional - the n-tips, binary, rooted, and ultrametric host tree to simulate the alignments on it (by default a pure-birth host tree is simulated)
n	optional - a positive integer value representing the number of tips of the host tree to simulate (need to be provided if you do not already provide a host tree)
mu	a positive value representing the simulated substitution rate.
N	a positive integer value representing the total number of nucleotides (variable and unvariable nucleotides)
proportion_variant	a value between 0 and 1 representing the proportion of variable nucleotides among N
model	Several ways to simulate host-switches on the host phylogeny are available: "model=uniform" simulates host-switches uniformly on the host phylogeny, "model=temporal" simulates host-switches preferentially at a given time in the past, "model=host_dependent" simulates host-switches preferentially between closely related hosts lineages, and "model=geo_dependent" simulates host-switches preferentially between host lineages sharing the same geographic area. (default="uniform").
mean	optional – for "model=temporal", the host-switches are simulated at a particular time in the past. The temporal distribution of host-switches follows a normal distribution centered on the "mean" and with a standard deviation "sd". "mean" must be a number between 0 and 1 indicating at what time in the past host-switches are more likely: "means=0" simulates host-switches closed to the root of the host phylogeny, whereas "means=1" simulates host-switches closed to the tips.
sd	optional – for "model=temporal", the host-switches are simulated at a particular time in the past. The temporal distribution of host-switches follows a normal distribution centered on the "mean" and with a standard deviation "sd" ("sd" must be positive).
host_signal	optional – for "model= host_dependent", the host-switches are preferentially simulated between host lineages that are closely related. "host_signal" must be a positive number. If "host_signal=0", switches are uniformly distributed, whereas the phylogenetically preference increases with the "host_signal" value.
geo_signal	optional – for "model=geo_dependent", the host-switches are preferentially simulated between host lineages sharing the same geographic area. "geo_signal" must be between 0 and 1, it indicates the probability for a switch to occur between areas divided by the probability for a switch to occur within an area. If "geo_signal"=1, switches are uniformly distributed, whereas if "geo_signal"=0, switches can only happen between host lineages sharing the same geographic area.
stochastic_map	optional – for "model=geo_dependent", the host-switches are preferentially simulated between host lineages sharing the same geographic area. "stochastic_map" represents a stochastic mapping that indicates the ancestral biogeography of the host lineages (it indicates which host lineages occupied which geographic areas at a given time, and thus indicates which lineages are sharing the same vs. different areas). "stochastic_map" must be an A object of class "simmap" (R-package phytools, Revell, 2012).
delta	rate of duplication (by default no duplication are simulated: delta=0)
seed	a seed to assure the reproducibility
nb_cores	a number of cores to run the analyses
path	optional - the path to the working directory
...	optional - other arguments to be passed.

Details

If you provide a binary rooted ultrametric host tree, it must be in Newick format (.tre).

Value

The function outputs microbial nucleotidic alignments according to different models of Host-Microbiota evolution. Data are well-formatted to then directly run HOME (HOME_model).

Author(s)

Benoit Perez-Lamarque

References

Perez-Lamarque B, Morlon H (2019). Characterizing symbiont inheritance during host-microbiota evolution: Application to the great apes gut microbiota. Molecular Ecology Resources 19:1659-1671.

See Also

HOME_model simulate_alignment

Examples

# Simulate 3 microbial alignments on a host tree
# (1 is vertically transmitted, 1 is transmitted with 5 host-switches,
#  and 1 is environmentally acquired)

sim_microbiota(name="example_simulation", name_index=c("Simul_1","Simul_2","Simul_3"),
               simul=c(0,5,"indep"), n=10, mu=1,
               N=300, proportion_variant=0.1)


# Run HOME (parameters estimations and model selections)

#HOME_model(name="example_simulation", name_index=c("Simul_1","Simul_2","Simul_3"),
#nb_tree=1000, lambda=seq(1,25), empirical=FALSE, nb_random=10)


# Non-uniform types of host-switches:

# host-phylogenetic dependency
# Simulate 3 microbial alignments on a host tree
# (1 is vertically transmitted,
#  1 is transmitted with 5 host-switches (with host-phylogenetic dependency),
#  and 1 is environmentally acquired)

#sim_microbiota(name="example_simulation", name_index=c("Simul_1","Simul_2","Simul_3"),
#               simul=c(0,5,"indep"), n=10, mu=1,
#               N=300, proportion_variant=0.1, model="host_dependent",
#               host_signal = 10)

# geographic dependency
# Simulate 3 microbial alignments on a host tree
# (1 is vertically transmitted,
#  1 is transmitted with 5 host-switches (with geographic dependency),
#  and 1 is environmentally acquired)

set.seed(3)
host_tree <- phytools::pbtree(n=20) # simulate host phylogeny

# simulate area distribution of the host lineages
geo <- sample(size = 20,c("Area_1","Area_2","Area_3","Area_4"), replace=TRUE)
names(geo) <- host_tree$tip.label

#stochastic_map <- phytools::make.simmap(host_tree, geo, model="SYM", nsim=1)

#sim_microbiota(name="example_simulation", name_index=c("Simul_1","Simul_2","Simul_3"),
#               simul=c(0,5,"indep"), mu=1, provided_tree = host_tree,
#               N=300, proportion_variant=0.1, model="geo_dependent",
#               geo_signal = 0.25, stochastic_map = stochastic_map)


# temporal dependency
# Simulate 3 microbial alignments on a host tree
# (1 is vertically transmitted,
#  1 is transmitted with 5 host-switches (with temporal dependency),
#  and 1 is environmentally acquired)

#sim_microbiota(name = "example_simulation", name_index = c("Simul_1","Simul_2","Simul_3"),
#               simul = c(0,5,"indep"), n = 10, mu = 1,
#               N = 300, proportion_variant = 0.1, model = "temporal",
#               mean = 0.5, sd = 0.1)

BPerezLamarque/HOME documentation built on May 17, 2023, 7:02 a.m.