A simple pre-disease state prediction method based on variations of gene vector features.

Area of Science:

• Bioinformatics
• Computational Biology
• Systems Biology

Background:

• Disease progression involves normal, pre-disease, and disease states.
• The pre-disease state is a critical juncture for disease deterioration.
• Early prediction of pre-disease states can facilitate timely prevention and treatment.

Purpose of the Study:

• To develop a novel method for predicting pre-disease states.
• To utilize gene regulatory network information for pre-disease prediction.
• To enhance early disease detection and intervention strategies.

Main Methods:

• Representing gene expression as gene vectors to capture their states.
• Applying gene vector features within a gene regulatory network framework.
• Developing a novel pre-disease prediction algorithm based on vector representations.

Main Results:

• Successfully predicted pre-disease states in an influenza virus infection dataset.
• Identified pre-disease related genes that are highly associated and enriched in relevant pathways.
• Demonstrated improved computational efficiency compared to existing methods.

Conclusions:

• The proposed gene vector method effectively predicts pre-disease states.
• The method aids in identifying biologically relevant genes for disease states.
• This approach offers a computationally efficient tool for early disease detection.