Daniel R. Olson

NEAR improves homology detection speed and accuracy by using neural embeddings trained with contrastive learning, outperforming existing methods as a pre-filter for protein sequence annotation.

The paper introduces ULTRA, a tool for identifying and annotating repetitive genomic sequences, providing reliable coverage and interpretable statistics for use in annotation pipelines.

The paper presents the first de novo repeat discovery and annotation of a complete human reference genome, identifying novel satellite arrays and characterizing new classes of complex repeats.

The paper presents a comprehensive analysis of fully assembled human centromeres, revealing their architecture, variation, and evolutionary dynamics at high resolution.

PolyA adjudicates competing sequence annotations by computing confidence estimates and identifying optimal alignments with rearrangements.