Identifying critical states of hepatocellular carcinoma based on landscape dynamic network biomarkers. (April 2020)

• Record Type: Journal Article
• Authors: Sun, Yichen; Zhao, Hongqian; Wu, Min; Xu, Junhua; Zhu, Shanshan; Gao, Jie
• Is Part Of: Computational biology and chemistry. Volume 85(2020)
• Publication Date: 2020-04
• Subjects: Hepatocellular carcinoma (HCC) -- Critical state -- Landscape dynamic network biomarkers (L-DNB) -- Statistical characteristic; Chemistry -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; Biochemistry -- Data processing; Biology -- Data processing; Molecular biology -- Data processing; Periodicals; Electronic journals; 542.85
• DOI: 10.1016/j.compbiolchem.2020.107202
• Languages: English
• ISSNs: 1476-9271

Abstract

Graphical abstract: This paper constructs a novel method based on landscape dynamic network biomarkers (L-DNB), which aims to detect early warning signals from cirrhosis state to very advanced HCC state in individual patient. The results can provide scientific advice for early warning indicators and optimal treatment time for HCC. Highlights: Our algorithm improves the original landscpe-DNB method with strong clinical applicability. We identify the critical states of HCC based on the characteristics of the distribution of the sample population. In our study, low-grade dysplastic and high-grade dysplastic were identified as critical states of HCC. Abstract: Hepatocellular carcinoma (HCC) is the major histological form of primary liver cancer. It has usually reached the disease state once the patient is diagnosed since there are no specific symptoms in the early stages of HCC. This fact increases the difficulty of curing HCC. Recently, quantities of evidence have shown that many mathematical methods (such as dynamic network biomarkers, DNB) can be used to detect critical states or tipping points of complex diseases. However, it is difficult to apply the DNB theory to the clinic since multiple samples are generally unavailable for individual patient. This paper constructs a novel method based on landscape dynamic network biomarkers (L-DNB), which aims to detect early warning signals from cirrhosis state to very advanced HCC state in individual patient. The selected dataset contains multiple samples for each HCC state. A score that indicates the disease characteristics is calculated for each sample by RNA-seq data, and several scores constitute a distribution in the same state. Quantifying the statistical characteristics of these distributions and determining that low-grade dysplastic and high-grade dysplastic are the critical states of HCC. These results can provide scientific advice for early warning indicators and optimal treatment time for HCC.