Enhancing RNA-seq bias mitigation with the Gaussian self-benchmarking framework: towards unbiased sequencing data

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• RNA sequencing is crucial for cellular RNA analysis but prone to data-distorting biases.
• Existing bias correction methods are empirical, address biases individually, and have limited effectiveness.

Purpose of the Study:

• Introduce the Gaussian Self-Benchmarking (GSB) framework for simultaneous mitigation of multiple RNA sequencing biases.
• Leverage natural guanine-cytosine (GC) content distribution patterns for a theoretical approach to bias correction.

Main Methods:

• Organize k-mers based on GC content according to a theoretical model.
• Apply a Gaussian model for alignment to ensure empirical data matches theoretical distributions.
• Utilize pre-determined GC content distribution parameters (mean, standard deviation) for precise RNA sample representation.

Main Results:

• GSB demonstrated superior bias mitigation compared to existing methods.
• Effectively addressed individual and co-existing biases in synthetic and human RNA samples.
• Improved accuracy and reliability of RNA sequencing data.

Conclusions:

• GSB is a significant advancement for multi-bias mitigation in RNA sequencing.
• Establishes a new standard for unbiased RNA sequencing results, independent of dataset flaws.
• Enhances RNA study reliability, advancing genetic disease research and targeted treatment development.