shuangj00 / HARMONIES

A Hybrid Approach for Microbiome Networks Inference via Exploiting Sparsity
5 stars 5 forks source link

Explanation of Simulated datasets #5

Open MonicaSteffi opened 3 years ago

MonicaSteffi commented 3 years ago

Dear all,

Sorry for this basic question. I need a simulation datasets with ground truth to test the performance of different methods including HARMONIES, MDINE, SPIEC-EASY. Luckily I found your R package and I tried it myself. you have provided simulated RData for different number of samples and features.

I have trouble in understanding the structure of simulated RData. Each RData has 50 list and in each list (data), there are six sublists including Y, A, A_graph, Phi, Sigma, mu_loga, ZI, Ztrue, Ztrue_pct., pcor, omega, Corr, sigma and adjcent mat.

https://drive.google.com/file/d/1uhU9tA4e1vKtYiMvxBmI-tOKOnknY5sb/view?usp=sharing

Here, Y represents the simulated count matrix (without any normalization), mu.loga: the underlying multivariate normal data that used as a parameter in the DM data generation and Sigma the variance-covariance matrix of the multivariate normal data.

But what does others represent: like A, A_graph, etc. I hope i can extract Y from simulated data and test the performance. Kindly correct me if I am wrong