ARCLID: Accurate and Robust Characterization of Long Insertions and Deletions in Genome
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Accurate detection of genomic structural variants (SVs) remains challenging, particularly for large events and at low sequencing coverage. We present ARCLID, a novel deep learning-based SV caller for PacBio HiFi data that treats SVs as objects in pileup images. Evaluated on diverse real and synthetic datasets across varying coverage levels, ARCLID demonstrates reliable and consistent accuracy, even for SVs larger than 1 kbp. Crucially, it maintains high performance on samples with lower sequencing depths (10X, 5X). This capability to preserve accuracy at lower depths provides a significant practical advantage for cost-constrained projects, enabling robust SV discovery without requiring deep sequencing. ARCLID represents a promising advancement toward more accessible and efficient long-read SV analysis for diverse research applications.
