A Data-Distribution and Successive Spline Points based discretization approach for evolving gene regulatory networks...

Area of Science:

• Systems Biology
• Computational Biology
• Bioinformatics

Background:

• Gene regulatory network (GRN) inference is crucial in Systems Biology.
• Boolean networks offer a simplified GRN modeling approach but require binarized gene expression data (GED).
• Single-cell RNA sequencing (scRNA-Seq) presents unique challenges, including high zero-data frequency, complicating traditional GED discretization.

Purpose of the Study:

• To propose a novel discretization method, Distribution and Successive Spline Points Discretization (DSSPD), tailored for scRNA-Seq time-series data.
• To integrate DSSPD with Cartesian Genetic Programming (CGP) for enhanced GRN inference.
• To evaluate the performance of the proposed DSSPD-CGP approach against standard methods and state-of-the-art algorithms.

Main Methods:

• Development of the Distribution and Successive Spline Points Discretization (DSSPD) method, incorporating data distribution and preprocessing steps.
• Application of Cartesian Genetic Programming (CGP) for GRN inference utilizing DSSPD-processed data.
• Comparative analysis of DSSPD-CGP against CGP with standard data handling and five established algorithms using curated and experimental datasets.

Main Results:

• The proposed DSSPD method significantly improves GRN inference accuracy when used with CGP.
• DSSPD-CGP consistently outperformed standard CGP data handling across all tested scenarios.
• The DSSPD-CGP approach demonstrated superior performance compared to five state-of-the-art algorithms in the majority of evaluated cases.

Conclusions:

• DSSPD is an effective discretization strategy for scRNA-Seq time-series data, particularly addressing zero-inflation.
• The integration of DSSPD with CGP provides a robust framework for accurate gene regulatory network inference.
• This approach offers a significant advancement for systems biology research utilizing high-dimensional single-cell transcriptomic data.