I am a high school student, and this is a published paper that I wrote. If you want to read a shorter blog version of my work, go here: https://blog.liudaniel.com/n-grams-BK-trees.

The general problem is grouping similar DNA/RNA sequences based on something known as a Unique Molecular Identifier, and then collapsing those groups into consensus sequences. This helps estimate the number of unique sequences while efficiently accounting for errors in sequencing or PCR amplification.

If you have any questions, feel free to ask me!