[SS06] Data-based dynamical modeling of complex biological systems

[SS06-02] Mathematical AI for life science

*Luonan Chen¹ (1. School of Mathematical Sciences and School of AI, Shanghai Jiao Tong University (China))

Keywords:

AI,DNB,tipping point,time series,causality

In this talk, I will deliver the topics of mathemtical AI for life science, i.e. mAI4LS, including the topics of tipping point detection, time-series prediction prediction, causal inference for biological/disease processes, based on dynamical data science and AI. In particular, I will show the methodology for analyzing critical transitions of biological systems by DNB (dynamic network biomarker). There are critical transition phenomena during many of disease progressions. Such critical transitions are usually accompanied by catastrophic disease deterioration, such as cancer and diabetes. The prediction of such disease deterioration/transition is of great significance for disease prevention and treatment. However, predicting the disease deterioration for an individual is a difficult but highly practical problem. In this study, we presented a novel DNB/AI-based method, i.e., network information gain (NIG) method for predicting such a critical transition based on network flow entropy from omics data of each individual. NIG can not only efficiently predict disease deteriorations but also detect their dynamic network biomarkers (DNBs) on an individual basis, and further effectively identify potential therapeutic targets. The numerical simulation demonstrates the effectiveness of NIG and also the superiority over existing methods. Moreover, our method was validated by successfully detecting the critical states before the disease deterioration and also identifying their potential therapeutic targets, from various real omics datasets, i.e., influenza, bladder urothelial carcinoma (BLCA), cervical squamous cell carcinoma (CESC) and liver hepatocellular carcinoma (LIHC).