DivMigrate was run based on the script by Sundqvist et al. (2016), as described in the paper: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4842207/. To enhance the analysis for the study, I have used Python programming to independently run simulations and determine the best scenario for choosing an analysis.
The DivMigrate simulation in Python (pyMigSim) runs 1,000-10,000 replications to select the highest values, employing cosine distance calculation. The method computes the cosine distance between two vectors in 3D spatial using the SciPy module in Python.
To run the simulation:
- Execute migration using DivMigrate in RStudio (
python pyDivMigSim.py <R_script> <genpop_file> <method [nm|d|g]> <simulation>) - Determine the highest value of the simulation number using cosine distance and selecting all the candidate's simulation running (
python SIMcos_DivMigrate.py <method [nm|d|g]> <simulation>) - Compute statistics of
OLS (Ordinary least-squares)for all the best simulations to select the optimal one
Rstudio 4.2.2
r-diveRsity
python 3.8.15
numpy
pandas
seaborn
matplotlib
scipy
statsmodels
sklearn
mpl_toolkits
plotly
imageio
- All cosine distance in 3D plot
- The candidate simulations for choosing based on the highest similarities
- Plotting the regression from all dataset
- Choosing the final simulation between the candidate sims use OLS model
Results: Ordinary least squares
======================================================================
Model: OLS Adj. R-squared: -inf
Dependent Variable: const AIC: -3349.4865
Date: 2024-04-23 05:09 BIC: -3345.7028
No. Observations: 49 Log-Likelihood: 1676.7
Df Model: 1 F-statistic: -47.00
Df Residuals: 47 Prob (F-statistic): 1.00
R-squared: -inf Scale: 1.1565e-31
----------------------------------------------------------------------
Coef. Std.Err. t P>|t| [0.025 0.975]
----------------------------------------------------------------------
const 1.0000 0.0000 15882837056526484.0000 0.0000 1.0000 1.0000
sim_7660 0.0000 0.0000 2.4905 0.0163 0.0000 0.0000
sim_7444 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_4722 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_5246 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_9702 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_9716 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_4689 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_8344 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_8805 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_7160 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_5781 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_4431 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_6719 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_993 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_2465 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_8684 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_4309 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_3791 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_1943 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_9745 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_674 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_4027 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_7401 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_7294 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_7877 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_2144 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_7306 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_941 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_7792 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_899 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_8918 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_1347 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_6957 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_4968 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_6946 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
sim_5116 -0.0000 0.0000 -0.1868 0.8526 -0.0000 0.0000
sim_6169 -0.0000 0.0000 -0.1868 0.8526 -0.0000 0.0000
sim_7849 0.0000 0.0000 0.0623 0.9506 -0.0000 0.0000
----------------------------------------------------------------------
Omnibus: nan Durbin-Watson: 0.000
Prob(Omnibus): nan Jarque-Bera (JB): nan
Skew: nan Prob(JB): nan
Kurtosis: nan Condition No.: 13292482535527436
======================================================================
Notes:
[1] Standard Errors assume that the covariance matrix of the errors is
correctly specified.
[2] The smallest eigenvalue is 2.51e-30. This might indicate that
there are strong multicollinearity problems or that the design
matrix is singular.
###########################
Your final simulation after running OLS was:
Coefficients 8.673617e-18
Name: sim_7660, dtype: float64
Genome-wide SNP analyses reveal high gene flow of endemic Smallscale Croaker (Boesemania microlepis) in the Lower Mekong Basin (2024)
Quyen Dang Ha Vu (quyenvdh@ntu.edu.vn), Binh Thuy Dang (binhdt@ntu.edu.vn), Oanh Thi Truong (oanhtt@ntu.edu.vn), Sang Quang Tran (sangtq@ntu.edu.vn), Bird E. Christopher (chris.bird@tamucc.edu), Kent Carpenter (kcarpent@odu.edu), Biesack E. Ellen (ebies001@odu.edu), Ackiss S. Amanda (aacki002@odu.edu)
############################################
Tsang - Sang Tran Quang
PhD student
Molecular Bioinformatics
Nha Trang University
sangtq@ntu.edu.vn
Khanh Hoa, Vietnam
############################################