Paragraph

Paragraph genotypes structural variants (SVs) from short-read sequencing data using sequence graphs to improve detection and genotyping accuracy.

Key Features:

• Sequence Graph Modeling: Leverages sequence graphs to model structural variations, integrating sequence information and SV annotations.
• Graph-based Alignment: Aligns short-read sequencing data onto sequence graphs to support variant detection and genotyping.
• Enhanced Accuracy: Demonstrates superior accuracy in detecting and genotyping SVs when benchmarked against long-read SV calls.
• Scalability for Population Studies: Has been applied at scale to a cohort of 100 short-read sequenced samples from individuals of diverse ancestries.

Scientific Applications:

• Genomics Research: Enables more accurate detection and genotyping of SVs to study genomic variation and its implications in health and disease.
• Clinical Sequencing Pipelines: Provides precise SV genotyping for clinical diagnostic workflows that rely on short-read sequencing data.

Methodology:

Constructs sequence graphs that incorporate reference sequences and known SV annotations, and aligns short-read sequencing data onto these graphs to identify and genotype structural variations.