Reliability of Serum-Derived Connectome Indicators in Identifying Cirrhosis

Patients
with cirrhosis face a heightened risk of complications,
underscoring the importance of identification. We have developed a
Connectome strategy that combines metabolites with peptide spectral
matching (PSM) in proteomics to integrate metabolomics and proteomics,
identifying specific metabolites bound to blood proteins in cirrhosis
using open search proteomics methods. Analysis methods including Partial
Least Squares Discriminant Analysis (PLS-DA), Uniform Manifold Approximation
and Projection (UMAP), and hierarchical clustering were used to distinguish
significant differences among the Cirrhosis group, Chronic Hepatitis
B (CHB) group, and Healthy group. In this study, we identified 81
cirrhosis-associated connectomes and established an effective model
distinctly distinguishing cirrhosis from chronic hepatitis B and healthy
samples, confirmed by PLS-DA, hierarchical clustering analysis, and
UMAP analysis, and further validated using six new cirrhosis samples.
We established a Unified Indicator for Identifying cirrhosis, including
tyrosine, Unnamed_189.2, thiazolidine, etc., which not only enables
accurate identification of cirrhosis groups but was also further validated
using six new cirrhosis samples and extensively supported by other
cirrhosis research data (PXD035024). Our study reveals that characteristic
cirrhosis connectomes can reliably distinguish cirrhosis from CHB
and healthy groups. The established unified cirrhotic indicator facilitates
the identification of cirrhosis cases in both this study and additional
research data.