Abstract : The concept of dynamical network biomarker (DNB) was proposed to provide early-warning signals of diseases on the basis of co-dimension 1 local bifurcation in 2012, and then is widely used in various topics and fields of biology and medicine, e.g. dynamical analyses of biological processes in biology and disease prediction/early-diagnoses in medicine. The DNB is a novel type of biomarkers to identify the critical state during disease progression, which quantifies biological systems from a dynamical and network viewpoint, thus providing reliable information on early-warning signals before onset of complex diseases. Many DNB methods as well as applications have been developed. The DNB theory with big biological data is expected to lead to ultra-early precision and preventive medicine. This symposium addresses but not limited to the recent development of theory, methodology and application of DNB in a variety of scientific areas.
Abstract : As the opening talk of this minisymposium, I introduce the concept of DNB (Dynamical Network Biomarkers) theory, its experimental proof of concept and its possible applications to ultra-early precision medicine.
[05641] DNB-based intervention for ultra-early treatment
Format : Talk at Waseda University
Author(s) :
Jun-ichi Imura (Tokyo Institute of Technology)
Abstract : The process leading to the onset of disease may be understood as a rapid transition in the complex interactions between genes. This talk proposes a DNB-based intervention approach for preventive treatment just before such transitions, i.e., in the pre-disease state, by combining DNB theory with control theory to build a theory that estimates which nodes of the relevant gene expression network should be intervened and how they should be intervened.
[05664] Machine learning techinques meet complex dynamical systems: A few recent advances
Format : Talk at Waseda University
Author(s) :
Wei LIN (Fudan University)
Abstract : In this talk, we will review the relations between the machine learning techniques and the complex dynamical systems. Also we will introduce the recent advances achieved in this area, including the developed theory of reservoir computing (RC), the application of RC in system reconstruction and prediction, change-point detection, and tracking unstable periodic orbits, and noise-driven controlling in nonlinear dynamical systems. The talk will be concluded with a few perspective on the future research along the above direction.
[05646] Deciphering Key Causal Mechanisms for Guided Phenotype and State Transitions
Format : Talk at Waseda University
Author(s) :
Chengming Zhang (The University of Tokyo)
Abstract : Understanding and manipulating cell state transitions is pivotal in biology. We introduce CauFinder, an innovative causal decoupling framework leveraging neural networks to emulate biological systems' hierarchical control, pinpointing crucial causal control nodes. Through optimizing information flow and isolating causal elements in latent spaces, CauFinder offers concise, accurate, and causal insights into phenotype and state transitions.
[05648] Early warning signals for multistage transitions in tipping dynamics on networks
Format : Talk at Waseda University
Author(s) :
Neil G. MacLaren (State University of New York at Buffalo)
Prosenjit Kundu (State University of New York at Buffalo)
Naoki Masuda (State University of New York at Buffalo)
Abstract : Complex dynamical systems for which we want to anticipate sudden regime shifts often form a heterogeneous network. We propose methods to select sentinel nodes in a given network to construct informative early warning signals given that the network may be heterogeneous and show multistage transitions. We show that small subsets of nodes can anticipate transitions as well as or even better than using all the nodes under the proposed node selection method. Informative sentinel nodes depend on the direction of regime shifts, which we also highlight in the talk.
[05343] Change-point detection in temporal complex systems
Format : Talk at Waseda University
Author(s) :
Huanfei Ma (Soochow University)
Abstract : We develop a model-free approach, named temporal change-point detection (TCD), and integrate both dynamical and statistical methods to achieve accurate detection of the time instant at which a system changes its
internal structures. The proposed approach is able not only to detect the separate change points of the concerned systems without knowing, a priori, any information of the equations of the systems, but also to harvest all the change points emergent in a relatively high-frequency manner.
[05637] Alerting for the critical transition of complex systems
Format : Talk at Waseda University
Author(s) :
Rui Liu (South China University of Technology)
Abstract : It is a challenging task to accurately predict the future critical state of a short-term time-series. The major difficulty to solve such a task is the lack of the information, which typically results in the failure of most existing approaches due to the overfitting problem of the small sample size. To address this issue, we proposed a computing framework: auto-reservoir neural network, to efficiently and accurately make the multi-step-ahead prediction based on a short-term high-dimensional time-series. Different from traditional reservoir computing whose reservoir is an external dynamical system irrelevant to the target system, ARNN directly transforms the observed high-dimensional dynamics as its reservoir, which maps the high-dimensional/spatial data to the future temporal values of a target variable based on a spatiotemporal information (STI) transformation. Combining with the dynamic network biomarker (DNB), it is possible to detect the early-warning signal of critical transitions of real-world complex systems.
[05638] The algorithm and application of landscape-DNB in complex disease of single sample
Format : Talk at Waseda University
Author(s) :
Xiaoping Liu (Hangzhou Institute for Advanced Study)
Abstract : A new model-free method has been developed and termed the landscape dynamic network biomarker (l-DNB) methodology. The method is based on bifurcation theory, which can identify tipping points prior to serious disease deterioration using only single-sample omics data. Here, we show that l-DNB provides early-warning signals of disease deterioration on a single-sample basis and also detects critical genes or network biomarkers (i.e. DNB members) that promote the transition from normal to disease states. As a case study, l-DNB was used to predict severe influenza symptoms prior to the actual symptomatic appearance in influenza virus infections. The l-DNB approach was then also applied to three tumor disease datasets from the TCGA and was used to detect critical stages prior to tumor deterioration using an individual DNB for each patient. The individual DNBs were further used as individual biomarkers in the analysis of physiological data, which led to the identification of two biomarker types that were surprisingly effective in predicting the prognosis of tumors. The biomarkers can be considered as common biomarkers for cancer, wherein one indicates a poor prognosis and the other indicates a good prognosis.
Yong Wang (Academy of Mathematics and Systems Science, Chinese Academy of Sciences)
Abstract : In this talk I will introduce combinatorial regulon (cRegulon) to model the combinations among TFs, which can better characterize cell types and serves as the driving forces for cell state transitions. By leveraging rapidly accumulated single multi-omics data, we develop an optimization model to systematically infer cRegulons (i.e., the representative TF modules, their associated regulatory elements and target genes formed regulatory network).
[05640] DNB based network fluctuation and application to biology and medicine
Format : Talk at Waseda University
Author(s) :
Luonan Chen (Chinese Academy of Sciences)
Abstract : I will talk about the recent progress on the DNB methods as well as the applications to biology and medicine. By exploring the original DNB concept, i.e. critical collective fluctuation (CCF) of the observed variables, we developed a network flow entropy, which can quantify the CCF so as to detect the tipping point before the critical transition from one stable equilibrium to another. The applications include the tipping points of various diseases.
