BoolFilter: an R package for estimation and identification of partially-observed Boolean dynamical systems

Area of Science:

• Computational Biology
• Systems Biology
• Bioinformatics

Background:

• Gene regulatory networks control essential cellular processes.
• Boolean networks model gene interactions but rely on directly observable states.
• Transcriptomic data is indirectly and incompletely measured, posing challenges for Boolean network models.

Purpose of the Study:

• Introduce the R package BoolFilter for Partially-Observed Boolean Dynamical Systems (POBDS).
• Provide a computational tool for analyzing gene regulatory networks from noisy transcriptomic data.
• Address the limitations of existing Boolean network models in handling indirect measurements.

Main Methods:

• Implementation of the POBDS model in the BoolFilter R package.
• Utilizes exact and approximated (particle) filters for state and parameter estimation.
• Leverages network interface from the BoolNet R package for compatibility.

Main Results:

• BoolFilter enables estimation of Boolean states and network topology from time-series transcriptomic data.
• The package facilitates simulation of transcriptomic data based on Boolean network models.
• It handles uncertainty in measurement processes inherent in transcriptomic analysis.

Conclusions:

• BoolFilter offers a robust toolbox for the bioinformatics community.
• Provides state-of-the-art algorithms for simulating and identifying gene regulatory systems.
• Supports various expression technologies including cDNA microarrays, RNA-Seq, and cell imaging.