Yclon: Ultrafast clustering of B cell clones from high-throughput immunoglobulin repertoire sequencing data

Motivation Next-generation sequencing technologies revolutionize our understanding of immunoglobulin (Ig) profile in different immune states. Clonotyping (grouping Ig sequences into B cell clones) allows the investigation of the diversity of repertoires and how they change upon antigen exposure. Despite its importance, there is no consensus on the best method for clonotyping, and the methods developed for that are computationally intractable for large sequencing datasets. This is the case of Change-O, which compares sequences through hamming distance and uses hierarchical clustering for this task. Results We propose implementing an approach to identify B cell clones from Ig repertoire data, named Yclon, that makes an alignment-free comparison of the sequences, focusing on reducing the runtime and computer memory usage. Overall, we find that a hierarchical clustering approach grouped Ig sequences into B cell clones similarly to Change-O. However, we observed that Yclon was around 35 times faster and even was able to process larger than 2 million sequences Ig repertoire, which is a critical part of repertoire studies and enables understanding antibody repertoire structure and affinity maturation. Availability and implementation YClon is written in Python3 and is available on GitHub () Contact joaodgervasio{at}gmail.com, liza{at}icb.ufmg.br ### Competing Interest Statement The authors have declared no competing interest.