Reference flow: reducing reference bias using multiple population genomes.

Area of Science:

• Bioinformatics
• Genomics
• Computational Biology

Background:

• Standard sequencing analysis relies on linear reference genomes, which can introduce reference bias due to unaddressed genetic variation.
• Graph-based methods incorporate genetic variation but demand significant computational resources.

Purpose of the Study:

• To introduce a novel alignment method, reference flow, designed to enhance accuracy and mitigate reference bias in sequencing data analysis.
• To offer a computationally efficient alternative to existing methods for handling genetic variation.

Main Methods:

• The reference flow alignment method utilizes multiple population reference genomes.
• This approach aims to improve the accuracy of aligning sequencing reads and reduce reference bias.

Main Results:

• Reference flow achieves comparable accuracy and bias reduction to the graph aligner vg.
• Reference flow demonstrates significant computational advantages, using 14% of the memory and operating 5.5 times faster than vg.

Conclusions:

• Reference flow alignment presents an accurate and computationally efficient solution for analyzing sequencing data with genetic variation.
• This method offers a practical alternative for large-scale genomic studies where computational resources are a concern.