Density and Conservation Optimization of the Generalized Masked-Minimizer Sketching Scheme

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Minimizers and syncmers are established k-mer sketching techniques for sequence analysis.
• Minimizers prioritize coverage and minimal sketch size, while syncmers offer robustness to base substitutions.
• Existing methods struggle to optimize both coverage and robustness simultaneously, especially for parameterized syncmers.

Purpose of the Study:

• Introduce masked minimizers, a novel sketching scheme generalizing minimizers.
• Develop a practical algorithm to optimize masked minimizers for density and conservation.
• Enable simultaneous optimization of sketch compactness, spread, and substitution robustness.

Main Methods:

• Developed the masked minimizer scheme, analogous to parameterized syncmers generalizing syncmers.
• Extended existing minimizer optimization techniques to the masked minimizer framework.
• Evaluated the optimization algorithm on diverse benchmark genomes.

Main Results:

• The masked minimizer optimization algorithm successfully balances density and conservation.
• Achieved more compact, well-spread, and substitution-robust sketches compared to previous methods.
• Demonstrated improved performance on benchmark genomic datasets.

Conclusions:

• Masked minimizers provide a flexible and optimizable sketching approach for genomic data.
• This technique enhances efficiency and robustness in genomic analyses utilizing k-mer sketching.
• The developed algorithm facilitates more effective sequence data representation and analysis.