Pacybara: accurate long-read sequencing for barcoded mutagenized allelic libraries.

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Long-read sequencing offers advantages but is limited by higher error rates.
• Accurate barcode identification is crucial for sequencing mutagenized libraries in applications like multiplexed assays of variant effects (MAVEs).
• Existing pipelines struggle with sequencing errors and non-unique barcodes, hindering genotype-phenotype mapping.

Purpose of the Study:

• To develop a computational tool, Pacybara, to address challenges in long-read sequencing of barcoded mutant libraries.
• To improve the accuracy of associating barcodes with genotypes in MAVEs.
• To enhance the reliability of genotype-phenotype maps for clinical variant interpretation.

Main Methods:

• Pacybara clusters long reads based on barcode similarity, accounting for sequencing errors.
• The tool detects barcodes associated with multiple genotypes and identifies recombinant clones.
• Pacybara reduces false positive indel calls in sequencing data.

Main Results:

• Pacybara successfully handles inaccurate sequencing and non-unique barcodes in barcoded mutant libraries.
• The software accurately identifies and resolves issues related to barcode-genotype association.
• Demonstrated effectiveness in three example applications, improving MAVE data quality.

Conclusions:

• Pacybara enhances the accuracy and reliability of long-read sequencing for barcoded mutant libraries.
• The tool facilitates more robust genotype-phenotype mapping for clinical variant interpretation.
• Pacybara is freely available and implemented for use on high-performance computing clusters.