In [1]:

    

using DataFrames,JWAS,JWAS.Datasets


    

In [2]:

    

MTData   = Datasets.dataset("testMT","MTData.txt");
genofile = Datasets.dataset("testMT","genotype.txt");


    

In [3]:

    

;cat $MTData


    

    




y1   y2    trt
1.0  2.0    1
1.1  NA     1
0.9  1.9    2
1.2   1.7   2

In [4]:

    

df = readtable(MTData, separator = ' ')


    

    
Out[4]:




y1
y2
trt
1
1.0
2.0
1
2
1.1
NA
1
3
0.9
1.9
2
4
1.2
1.7
2

In [5]:

    

;cat $genofile


    

    




Animal,x1,x2,x3,x4,x5
S1,1.0,0.0,1.0,1.0,1.0
D1,2.0,0.0,2.0,2.0,1.0
O1,1.0,2.0,0.0,1.0,0.0
O3,0.0,0.0,2.0,1.0,1.0

In [6]:

    

R0=[1.0 0.5
    0.5 2.0];


    

In [7]:

    

models = "y1 = intercept + trt;
          y2 = intercept + trt"
mme = build_model(models,R0,df=10);


    

In [8]:

    

G=[1.0 0.5
    0.5 2.0];
add_markers(mme,genofile,G,separator=',',header=true);


    

    




5 markers on 4 individuals were added.

In [9]:

    

Pi=Dict([1.0; 1.0]=>0.25,[1.0; 0.0]=>0.25,[0.0; 1.0]=>0.25,[0.0; 0.0]=>0.25)
out = runMCMC(mme,df,Pi=Pi,methods="BayesC",missing_phenotypes=true,
chain_length=5000,output_samples_frequency=100);


    

    




Priors for marker effects covariance matrix were calculated from genetic covariance matrix and π.
Marker effects covariance matrix is 
[0.876712 0.876712
 0.876712 1.753425].


MCMC Information:
methods                                      BayesC
chain_length                                   5000
estimatePi                                    false
constraint                                    false
missing_phenotypes                             true
starting_value                                false
output_samples_frequency                        100
printout_frequency                             5001
update_priors_frequency                           0

Degree of freedom for hyper-parameters:
residual variances:                          10.000
iid random effect variances:                  4.000
polygenic effect variances:                   4.000
marker effect variances:                      4.000



running MCMC for BayesC...100%|█████████████████████████| Time: 0:00:02

In [15]:

    

Pi=Dict([1.0; 1.0]=>0.25,[1.0; 0.0]=>0.25,[0.0; 1.0]=>0.25,[0.0; 0.0]=>0.25)
out = runMCMC(mme,df,Pi=Pi,methods="BayesC",constraint=true,
missing_phenotypes=true,chain_length=1000,printout_frequency=1000);


    

    




MCMC Information:
methods                                      BayesC
chain_length                                   1000
estimatePi                                    false
constraint                                     true
missing_phenotypes                             true
starting_value                                false
output_samples_frequency                          0
printout_frequency                             1000
update_priors_frequency                           0

Degree of freedom for hyper-parameters:
residual variances:                          10.000
iid random effect variances:                  4.000
polygenic effect variances:                   4.000
marker effect variances:                      4.000



running MCMC for BayesC... 73%|██████████████████       |  ETA: 0:00:00
Posterior means at iteration: 1000
Residual covariance matrix: 
[0.382224 0.0
 0.0 2.121606]
Marker effects covariance matrix: 
[0.048908 0.057839
 0.057839 0.085165]

running MCMC for BayesC...100%|█████████████████████████| Time: 0:00:00

In [16]:

    

out = runMCMC(mme,df,Pi=Pi,methods="BayesC0",missing_phenotypes=true,
chain_length=5000,update_priors_frequency=1000);


    

    




MCMC Information:
methods                                     BayesC0
chain_length                                   5000
estimatePi                                    false
constraint                                    false
missing_phenotypes                             true
starting_value                                false
output_samples_frequency                          0
printout_frequency                             5001
update_priors_frequency                        1000

Degree of freedom for hyper-parameters:
residual variances:                          10.000
iid random effect variances:                  4.000
polygenic effect variances:                   4.000
marker effect variances:                      4.000



running MCMC for BayesC0... 13%|███                     |  ETA: 0:00:01
 Update priors from posteriors.
running MCMC for BayesC0... 40%|██████████              |  ETA: 0:00:00
 Update priors from posteriors.
running MCMC for BayesC0... 54%|█████████████           |  ETA: 0:00:00
 Update priors from posteriors.
running MCMC for BayesC0... 67%|████████████████        |  ETA: 0:00:00
 Update priors from posteriors.
running MCMC for BayesC0... 93%|██████████████████████  |  ETA: 0:00:00
 Update priors from posteriors.
running MCMC for BayesC0...100%|████████████████████████| Time: 0:00:01