Modelling gene regulation networks via multivariate adaptive splines

Area of Science:

• Genomics and Bioinformatics
• Molecular Biology
• Systems Biology

Background:

• Characterizing genome-wide transcriptional regulation networks is crucial following large-scale genome sequencing projects.
• Understanding the regulatory functions of transcription factor binding sites in eukaryotes is essential for deciphering gene expression control.

Purpose of the Study:

• To propose and validate a novel computational approach for modeling the regulatory roles of motifs in gene expression time series data.
• To identify known and novel regulatory motifs involved in transcriptome reprogramming using this new method.

Main Methods:

• Development of a novel approach based on multivariate adaptive splines (MAS).
• Application of the MAS approach to analyze two meiotic gene expression time series datasets.
• Modeling of individual motifs and coupled motif interactions within regulatory networks.

Main Results:

• Identification of well-documented and novel putative regulatory motifs involved in transcriptome reprogramming.
• Successful modeling of motifs that exert their regulatory effects independently and in conjunction with other motifs.
• Demonstration of the MAS approach's capability in deciphering complex transcriptional regulatory networks.

Conclusions:

• Multivariate adaptive splines offer a powerful tool for analyzing gene expression time series data.
• The proposed approach enhances the understanding of transcriptional regulatory networks in eukaryotes.
• This method can uncover both individual and interacting regulatory motifs crucial for gene expression control.