EasyCen: A Lightweight Framework for Centromere Localisation and Repeat-Organisation Profiling in...

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Centromere identification in telomere-to-telomere genomes is challenging due to rapidly evolving and non-conserved repetitive sequences.
• Existing methods often rely on repeat annotation or homology, limiting their applicability to diverse eukaryotic genomes.

Purpose of the Study:

• To develop a novel, lightweight, sequence-based framework called EasyCen for accurate centromere identification and repeat-architecture profiling.
• To overcome limitations of current methods by focusing on intrinsic positional features of repetitive DNA.

Main Methods:

• EasyCen utilizes a sequence-based approach, recognizing centromeres by recurrent positional patterns of repetitive DNA, independent of repeat annotation or homology.
• Incorporates a repeat-pair profiling module for characterizing internal repeat organization within centromeres.
• Benchmarked on *Arabidopsis thaliana* and *Mus musculus* genomes.

Main Results:

• EasyCen demonstrated high accuracy (>85% coordinate overlap) and significantly reduced runtime compared to existing methods.
• Cross-species analysis revealed a conserved 'beads-on-a-string' repeat pattern in human and mouse centromeres, linked to GC- and CpG-rich subdomains.
• The framework performs effectively without requiring pre-existing repeat libraries.

Conclusions:

• EasyCen provides an efficient and accurate solution for centromere identification and repeat profiling, particularly for large, repeat-rich, or non-model genomes.
• The findings suggest that despite rapid sequence evolution, certain organizational features of centromeric repeats may be conserved across eukaryotic lineages.