TideCluster: clustering and classification of tandem repeats for comparative and pangenomic analysis

Satellite DNA, composed of tandem arrays of relatively short repeated units, is a ubiquitous component of plant and animal genomes. Despite its biological importance, systematic investigation of satellite DNA remains challenging, in part because integrated bioinformatic tools for structured comparative analyses across individual arrays, chromosomes, and genomes are lacking. Here, we present TideCluster, a reference-independent pipeline that converts fragmented tandem repeat detections into genome-wide, interpretable repeat family catalogs. Starting from a genome assembly, TideCluster consolidates detected regions into arrays, clusters arrays by sequence similarity, reconstructs representative monomer consensus sequences, estimates locus-specific periodicity, identifies higher-order repeats, and generates GFF3 annotations. In addition to predefined visualizations, including the chromosomal distributions of satellite repeat families, TideCluster produces information-rich outputs that can be adapted for user-specific downstream analyses. We demonstrate its capabilities using the well-characterized genome of Arabidopsis thaliana and a set of Solanum species with highly diverse satellite repeat complements that vary markedly in abundance and have monomer lengths ranging from a few base pairs to more than 20 kb.