CP-WFABC v1.0 detects allele trajectories of changing selection from those of constant selection using ABC model choice, and jointly estimates the position of a change point as well as the strength of both corresponding selection coefficients (and dominance for diploid cases) using Wright-Fisher ABC methods
N population size (number of chromosomes: N
individuals for haploids and N/2 individuals for
diploids) (1000 by default)
sample_times vector of exact sampling times in generations
N_sample vector of sample sizes in number of chromosomes
min_freq data ascertainment of a minimum frequency
at one of the sampling time points (0 for no
condition and 1 to condition on fixation)
N_allele data frame of observed SNPs in row (with rownames
=SNP names, column=sampled numbers))
max_sims maximum number of simulations to do before giving
up (1 by default)
no_sim number of simulated datasets to be created (1e6
by default)
best_sim number of best simulations to be used for
estimation and model choice (1e3 by default)
set_seed reproducible numbers (TRUE by default)
post_graph Posterior densities of M0 and M1 (FALSE by
default)
post_2D_M1 2D posteriors of M1 estimates (s1&s2, s1&CP,
s2&CP) (FALSE by default)
h_fixed h to be fixed in diploid populations (TRUE by
default)
h_given h to be used if fixed (0.5 by default)
ploidy 1 for haploids, 2 for diploids
j number of SNP appearing at time t0 in the
population (given as observed initial frequency
in N_allele * population size, except if j<1
given as j=1)
t0 time where SNP appears in generations (same as
data given in N_allele)
s_start time in generation when selection starts (s=0
before s_start) (same as t0)
PDFs of prior graphs for simulated parameters
Text files of summary of results (SNP_name M1_posterior_BF M0_estimate_s1 (M0_estimate_h) M1_estimate_s1 M1_estimate_ s2 M1_estimate_CP (M1_estimate_h))
PDFs of posterior graphs of parameters of interest (if TRUE)
PDFs of 2D posterior graphs of parameters of interest (if TRUE)
simulates a Wright-Fisher trajectory with changing selection or constant selection}
arguments {N,t,fluc_t,j,t0,s1,s2,h,s_start,ploidy,N_sample,sample_times,max_sims}
detects allele trajectories of changing selection from those of constant selection using ABC model choice, and jointly estimates the position of a change point as well as the strength of both corresponding selection coefficients and dominance for a diploid population using Change-Point Wright-Fisher ABC methods}
arguments {N,h_fixed,h_given,sample_times,N_sample,N_allele,min_freq,max_sims,no_sim,best_sim,set_seed,post_graph,post_2D_M1}
detects allele trajectories of changing selection from those of constant selection using ABC model choice, and jointly estimates the position of a change point as well as the strength of both corresponding selection coefficients for a haploid population using Change-Point Wright-Fisher ABC methods}
arguments {N,sample_times,N_sample,N_allele,min_freq,max_sims,no_sim,best_sim,set_seed,post_graph,post_2D_M1}
Shim, H., Laurent, S., Matuszewski, S., Foll, M., Jeffrey D Jensen (in review) Detecting and quantifying changing selection intensities from time-sampled polymorphism data.
Foll, M.*, Shim, H.*, & Jeffrey D Jensen (2014b) WFABC: A Wright-Fisher ABC-Based Approach for Inferring Effective Population Sizes and Selection Coefficients from Time-Sampled Data. Molecular Ecology Resources, 1, 87-98.
Foll, M., Poh, Y.-P., Renzette, N., Ferrer-Admetlla, A., Bank, C., Shim, H., Malaspinas, A.S., Ewing, G., Liu, P., Wegmann, D., Caffrey, D.R., Zeldovich, K.B., Bolon, D.N., Wang, J.P., Kowalik, T.F., Schiffer, C.A., Finberg, R.W. & Jensen, J.D. (2014a) Influenza Virus Drug Resistance: A Time-Sampled Population Genetics Perspective. PLoS Genetics, 10, e1004185.
population genetics; fluctuating selection; change point analysis; time-sampled data; approximate Bayesian computation; Wright-Fisher model; experimental design
see Example_diploid_model.R
see Example_haploid_model.R
Hyunjin Shim [[email protected]]