Motif-Based Text Mining of Microbial Metagenome Redundancy Profiling Data for Disease Classification

Area of Science:

• Microbiology
• Bioinformatics
• Computational Biology

Background:

• 16S rRNA text data is crucial for classifying microbiota-associated diseases.
• Raw text data requires systematic processing for feature extraction.
• High-dimensional feature spaces from text data present classification challenges.

Purpose of the Study:

• To develop a novel algorithm for analyzing 16S rRNA text data.
• To address the challenges of high-dimensional feature spaces in microbiota analysis.
• To improve the classification accuracy of microbiota-associated diseases.

Main Methods:

• Introduction of the Phylogenetic Tree-Based Motif Finding (PMF) algorithm.
• Integration of phylogenetic rules and statistical indexes for classification.
• Application of retrieved motifs with common classification methods.

Main Results:

• PMF effectively reduces the dimensionality of large feature spaces from text data.
• The algorithm successfully discriminates between samples of pneumonia and dental caries.
• PMF demonstrates superior performance compared to existing classification methods.

Conclusions:

• Phylogenetic approaches are extended for supervised learning on microbiota text data.
• PMF enhances efficiency and reliability in analyzing high-dimensional text data.
• The method shows promise for discriminating pathological states like pneumonia and dental caries.