This repository provides the code for sparse learning of probabilistic graphical models to identify the significant compounds and record the results.
The code version V1.0
If you find it interesting and confused about the content, please contact me.
communication E-mail: [email protected]
According to the reference [1], we perform Neighbourhood Selection [2], Graphical Lasso [3], Scale free networks by reweighed L1 regularization (glasso-SF) [4], and Sparse PArtial Correlation Estimation (space) [5] to select important compounds in the file "code/2-compound-target_network.R".
The results are shown in the file of result, and we plot the figures in the file of Figure.
[1] Liu, Q., & Ihler, A. (2011, June). Learning scale free networks by reweighted l1 regularization. In Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics (pp. 40-48).
[2] Meinshausen N, Bühlmann P. Variable selection and high-dimensional graphs with the lasso[J]. Annals of Statistics, 2006, 34: 1436-1462.
[3] Yuan M, Lin Y. Model selection and estimation in the Gaussian graphical model[J]. Biometrika, 2007, 94(1): 19-35.
[4] Peng J, Wang P, Zhou N, et al. Partial correlation estimation by joint sparse regression models[J]. Journal of the American Statistical Association, 2009, 104(486): 735-746.