Feature Selection of OMIC Data by Ensemble Swarm Intelligence Based Approaches.

Area of Science:

• Bioinformatics
• Computational Biology
• Machine Learning in Omics

Background:

• Omics datasets are characterized by high dimensionality.
• Complex interconnections among omics features pose challenges in linking them to biological traits.

Purpose of the Study:

• To develop an efficient method for identifying key biomarkers from high-dimensional omics data.
• To reduce feature dimensions effectively.
• To achieve accurate classification without extensive computational resources.

Main Methods:

• Ensemble swarm intelligence-based approaches.
• An end-to-end method relying solely on algorithmic rules.
• No presets for feature filtering, such as the number of features.

Main Results:

• Efficient identification of key biomarkers.
• Effective reduction of feature dimensions.
• Good classification accuracy achieved.
• Low consumption of computing resources.

Conclusions:

• Ensemble swarm intelligence offers an efficient solution for analyzing complex omics data.
• The proposed method simplifies biomarker discovery and feature selection.
• This approach is computationally efficient and accurate for biological trait association.