Decision tree-based method for integrating gene expression, demographic, and clinical data to determine disease...

Area of Science:

• Computational biology
• Genomics
• Systems biology

Background:

• Complex diseases present diagnostic and therapeutic challenges due to multifactorial etiology.
• Large datasets enable data-driven identification of distinct disease subtypes (endotypes).
• Challenges remain in segregating individuals into subtypes and understanding their distinct biological mechanisms.

Purpose of the Study:

• To develop and validate a novel method for defining data-driven disease endotypes.
• To improve the segregation of individuals into distinct disease subtypes.
• To facilitate the discovery of novel biological mechanisms underlying complex diseases.

Main Methods:

• A multi-step decision tree-based method was developed for endotype definition.
• The method integrates gene expression, clinical covariates, and disease indicators.
• Childhood asthma was used as a case study to evaluate the method's performance.

Main Results:

• The multi-step decision tree method outperformed alternative approaches like Student's t-test, single domain clustering, and Modk-prototypes.
• The novel method achieved superior segregation of asthmatics and non-asthmatics.
• The method provides clear identification of genes and clinical covariates distinguishing the groups.

Conclusions:

• The developed method enhances the understanding of complex diseases by enabling data-driven endotype discovery.
• This approach facilitates the identification of new disease mechanisms.
• It complements existing efforts and can link genetic and exposomic data for a comprehensive understanding of disease determinants.