Identifying Critical State of Complex Diseases by Single-Sample-Based Hidden Markov Model

The progression of complex diseases is generally divided as a normal state, a pre-disease state or tipping point, and a disease state. Developing individual-specific method that can identify the pre-disease state just before a catastrophic deterioration, is critical for patients with complex diseases. However, with only a case sample, it is challenging to detect a pre-disease state which has little significant differences comparing with a normal state in terms of phenotypes and gene expressions. In this study, by regarding the tipping point as the end point of a stationary Markov process, we proposed a single-sample-based hidden Markov model (HMM) approach to explore the dynamical differences between a normal and a pre-disease states, and thus can signal the upcoming critical transition immediately after a pre-disease state. Using this method, we identified the pre-disease state or tipping point in a numerical simulation and two real datasets including stomach adenocarcinoma and influenza infection, which demonstrate the effectiveness of the method.

Gene expression ranking change based single sample pre-disease state detection

Introduction

To prevent disease, it is of great importance to detect the critical point (pre-disease state) when the biological system abruptly transforms from normal to disease state. However, rapid and accurate pre-disease state detection is still a challenge when there is only a single sample available. The state transition of the biological system is driven by the variation in regulations between genes.

Methods

In this study, we propose a rapid single-sample pre-disease state-identifying method based on the change in gene expression ranking, which can reflect the coordinated shifts between genes, that is, S-PCR. The R codes of S-PCR can be accessed at https://github.com/ZhenshenBao/S-PCR.

Results

This model-free method is validated by the successful identification of pre-disease state for both simulated and five real datasets. The functional analyses of the pre-disease state-related genes identified by S-PCR also demonstrate the effectiveness of this computational approach. Furthermore, the time efficiency of S-PCR is much better than that of its peers.

Discussion

Hence, the proposed S-PCR approach holds immense potential for clinical applications in personalized disease diagnosis.