With the increasing availability of microbiome 16S data, network estimation has become a useful approach to studying the interactions between Network estimation on a set of variables is frequently explored using graphical models, in which the relationship between two variables is modeled via their conditional dependency given the other variables. In recent years, various methods for sparse inverse covariance estimation have been proposed to estimate graphical models in the high-dimensional setting, including graphical lasso. However, current methods do not address the compositional count nature of microbiome data, where abundances of microbial taxa are not directly measured, but are reflected by the observed counts in an error-prone manner. Adding to the challenge is that the sum of the counts within each sample, termed “sequencing depth”, is an experimental technicality, which carries no scientific information but can vary drastically across samples. To address these issues, we develop a new approach to network estimation, which models the microbiome data using a multinomial log-normal distribution with the finite sequencing depth explicitly incorporated. We propose to improve the empirical covariance estimator via a computationally simple procedure that corrects the bias arising from the heterogeneity in sequencing depth. We then build our inverse covariance estimator on graphical lasso. We will show the advantage of our method in comparison to current approaches for inverse covariance estimation under a variety of simulation scenarios. We will also illustrate the use of our method in an application to in human microbiome data set.