Bayesian state space models for dynamic genetic network construction across multiple tissues

Abstract

Construction of gene-gene interaction networks and potential pathways is a challenging and important problem in genomic research for complex diseases while estimating the dynamic changes of the temporal correlations and non-stationarity are the keys in this process. In this paper, we develop dynamic state space models with hierarchical Bayesian settings to tackle this challenge for inferring the dynamic profiles and genetic networks associated with disease treatments. We treat both the stochastic transition matrix and the observation matrix time-variant and include temporal correlation structures in the covariance matrix estimations in the multivariate Bayesian state space models. The unevenly spaced short time courses with unseen time points are treated as hidden state variables. Hierarchical Bayesian approaches with various prior and hyper-prior models with Monte Carlo Markov Chain and Gibbs sampling algorithms are used to estimate the model parameters and the hidden state variables. We apply the proposed Hierarchical Bayesian state space models to multiple tissues (liver, skeletal muscle, and kidney) Affymetrix time course data sets following corticosteroid (CS) drug administration. Both simulation and real data analysis results show that the genomic changes over time and gene-gene interaction in response to CS treatment can be well captured by the proposed models. The proposed dynamic Hierarchical Bayesian state space modeling approaches could be expanded and applied to other large scale genomic data, such as next generation sequence (NGS) combined with real time and time varying electronic health record (EHR) for more comprehensive and robust systematic and network based analysis in order to transform big biomedical data into predictions and diagnostics for precision medicine and personalized healthcare with better decision making and patient outcomes.

Computational dynamic approaches for temporal omics data with applications to systems medicine

“…The probability and confidence measures play important roles in omic temporal data not only due to the variations, high noise levels and experimental errors resident in the experiments but also the stochastic nature involved in the biological process. The Bayesian paradigm is very well suited for examining these features and other properties in the temporal data, such as highly correlated inputs (genes, time points) and phenotypes, missing data, and small sample size [ 50 – 56 ]. In Bayesian models, the parameters are assumed to be random variables and they are associated with some probability distribution, and the posterior probability of these parameters can be expressed as marginal distribution of those remaining parameters.…”

Temporal Dynamic Methods for Bulk RNA-Seq Time Series Data

“…Due to the lack of holistic comparative studies of various dynamic methods, there will be no universally or widely accepted or best methods available for different scenarios under time course experimental designs. Therefore, we recommend the development of a complete analytical pipeline to better characterize dynamic changes and to reduce misleading results (Figure 1) with detailed descriptions for each step) [1,6,10,15,17,19,20,23,25,28,33,44,46,47,[49][50][51][53][54][55]58,59,62,67,68,71,72,[74][75][76]81,[89][90][91][92][93][94][95][96][97][98][99][100][101][102][103].…”

“…(ii) GRNs in traditional microarrays: Time-varying multivariate state space model (SSM) tool [99] is designed for a short series of time course expression profiles where measurements from samples are collected between sparse and uneven time points. With the combination of Hidden Makov model and Dynamic Bayesian Network, HMDBN tool [95] is proposed by incorporating the non-stationary DBNs in which the structure and parameters are not fixed over a series of time points.…”