Discrimination of Etiologically Different Cholestasis by Modeling Proteomics Datasets.

Area of Science:

• Hepatology
• Proteomics
• Bioinformatics

Background:

• Cholestasis, characterized by impaired bile flow, presents diverse etiologies and symptoms.
• Accurate patient stratification is crucial for developing targeted therapies for cholestasis.
• Understanding the molecular basis of cholestasis progression is essential for improving patient care.

Purpose of the Study:

• To analyze the liver proteome of cholestatic patients to identify molecular differences.
• To develop a method for stratifying cholestasis patients based on proteomic profiles.
• To discover a protein panel capable of distinguishing between different cholestasis types.

Main Methods:

• Proteomic analysis of liver tissue from cholestatic patients and controls.
• Identification and quantification of liver proteins.
• Application of machine learning algorithms for patient stratification and biomarker discovery.

Main Results:

• Identified and quantified 7161 proteins, with 263 differentially expressed in cholestasis.
• Discovered deregulated cellular processes contributing to cholestasis progression.
• Developed a 20-protein panel using machine learning that accurately segregates different cholestasis types.

Conclusions:

• Proteomics combined with machine learning offers insights into cholestasis molecular mechanisms.
• The identified protein panel can discriminate between various cholestasis etiologies.
• This approach supports the development of precision medicine for cholestasis treatment.