The number of origins

Estimation of the number of origin of populaiton with spatial structure for inferring the effective population size

Flowchart

all_pkg.jl #before start: download all the packages + all questions summarized

1_*.jl, 2_*.jl: Non-spatial model

1_*.jl #multinomial distribution for genetic drift (Non-spatial model)

• 1_1allele_frequency.jl # the process of selection, genetic drift and mutation
  a. selection (adjust the frequency entering genetic drift)
  b. genetic drift (multinomial distribution) c. mutation (Poisson distribution)
• 1_2num_origin.jl # count the existant allele (frequency not=0) along the generations

test file:

• 1_1plot.jl #for testing whether 1_1allele_frequency.jl function well
• 1_3multinomial_plot.jl # test 1_2*.jl and plot

2_*.jl #Gaussian distribution for population size > 10^6 (Non-spatial model)

• 2_1allele_var_mtx.jl #covaraiance matrix
• 2_2gaussian_multinomial.jl #multivariate normal distribution (MvNormal) incorporating covariance
• 2_3allele_frequency_gaussian.jl # same process as 1_allele_frequency.jl, except replacing multinomial sampling for genetic drift by Gaussian approximation
• 2_4num_origin_gaussian.jl #count the existant allele test file:
  • 2_5gaussian_test.jl #test 2_4*.jl and plot

3_*.jl: One-dimensional model with local migration merely

   # a. use 1_allele_frequency.jl or 2_3allele_frequency_gaussian.jl  
   # b. local migration:  
          # total migration cases of each deme (Poisson sampling)--leaving  
           # dispersal direction of cases (Binomial: left or right)  
            # add the immigrants to the destination deme--entering  

3_1*.jl allele frequency

• 3_1_1track_cons.jl
  # trial: migration is tracked in individuals along with generations
• 3_1_2notrack_cons.jl #employed version for later series of functions, assumming #constant deme population size, migration only #change the frequeny of each advantageous haplotype
• 3_1_2notrack_cons_sparse.jl # sparse matrix based on 3_1_2notrack_cons.jl but for large data--more demes
• 3_1_2notrack_cons_t-1.jl # trial: mutation and migration taking the frequency of last generation as input of sampling instead of the #output of last process like Pac-Man
• 3_1_3notrack_osci.jl # employed oscillating population size

3_2*.jl Sampling 3_1*.jl

• 3_2_1d_tsamp.jl #time-series for constant population size
• 3_2_1d_tsamp_osci.jl #time-series for oscillating population size
• 3_3_1d_fixsamp.jl #collect the generation and asymptotic number of origin at fixation
• 3_3_1d_fixsamp_osci.jl # collect for oscillating population size
• 3_3_test.jl #check 3_*.jl series function right or not

3_4*.jl: compare parameters and plot

• 3_4_figjoejyn_series.jl # aligning the output from different parameters for later comparison
• 3_4_figjoejyn_series_threads.jl #multi-threads for speeding up (not for large data)
• 3_4_figjoejyn_series_test.jl #later comparison by plotting

4_*.jl: 1d model with migration in Student's t distribution (local + non-local)

• 4_1_1nonlocal_frequency.jl
  - selection + genetic drift + mutation
  - migration:
  - total migrants of each deme (Poisson sampling)
  - dispersal direction and how many deme the individual jump accross (t distribution sampling)
• 4_1_1nonlocal_frequency_sparse.jl # sparse matrix for large data--more demes
• 4_2_1d_tsamp.jl #time-series sampling of existant haplotype
• 4_3_1d_fixsamp.jl #collect the time to fixation and asymptotic number of origins
• 4_4_figjoejyn_series.jl #function collection for output with different parameters
• 4_4_figjoejyn_series_threads.jl #multi-thread added compared to 4_4_figjoejyn_series.jl

interest in comparison

• compare_nu.jl #compare the degree of freedom in sampling migration in t distribution

• 4and3_largenDemes.jl #large deme (nDemes = 500) #4 represent model with migration in t distribution #3 represent model with migration in binomial distribution

• 4_series_test_new.jl #all else

• test_sampling_distribution.jl #plotting different migration distribution

• time_cost.xlsx #time for simulation corresponding to the comparison and parameters applied #4_* represent in model with migration in t distribution #3_* represent in model with migration in binomial distribution