InvertypeR

InvertypeR performs discovery, genotyping, and phasing of genomic inversions from Single Cell Strand-seq data using a Bayesian binomial model on fixed genomic coordinates to increase power for detecting inversions smaller than 10 Kb.

Key Features:

• Bayesian inversion genotyping: Implements a Bayesian binomial model to genotype inversions, with validation on trios from the Human Genome Structural Variation Consortium showing reduction of Mendelian discordance from 6.3% to 0.5%.
• Automated detection and phasing: Automates inversion discovery, genotyping, and phasing directly from Single Cell Strand-seq data without manual coordinate calling.
• Fixed-coordinate analysis of small inversions: Operates on fixed genomic coordinates to increase statistical power for detecting inversions smaller than 10 Kb.
• Coordinate-flexible input: Accepts published inversion coordinates, predicted inversion hotspots (n = 3701), and coordinates identified by conventional methods for targeted analysis.
• Novel inversion discovery: Has genotyped 66 inversions that were previously unreported in the analyzed trios.

Scientific Applications:

• Genetic studies of phenotype and disease: Enables analysis of the contribution of genomic inversions to phenotypic variation, genome instability, and human disease.

Methodology:

Processes Single Cell Strand-seq data using an automated pipeline that applies a Bayesian binomial model on fixed genomic coordinates to perform inversion discovery, genotyping, and phasing.