LoMA

LoMA reconstructs localized assemblies from long-read sequencing data to resolve repetitive regions in human genomes at single-base resolution.

Key Features:

• Highly Accurate Consensus Sequences: Constructs consensus sequences from long reads using minimap2 for alignment, MAFFT for multiple sequence alignment, and a proprietary algorithm for classifying diploid haplotypes based on structural variants and consensus sequences.
• Error Reduction: Reduces sequencing error rates from over 8% in raw long-read data to below 0.3% in consensus sequences.
• Targeted Region Analysis: Defines target regions based on mapping patterns to focus assembly and analysis on specific genomic loci.
• Comprehensive Insertion Cataloging: Identified thousands of insertions in human samples NA18943 and NA19240, with approximately 80% derived from tandem repeats and transposable elements, and detected processed pseudogenes, insertions within transposable elements, and insertions >10 kbp.
• Insights into Genomic Mechanisms: Provides evidence linking short tandem duplications, gene expression, and transposons to mechanisms underlying genomic insertions.

Scientific Applications:

• Structural Variant Discovery: Resolve true structural variations in human genomes using high-accuracy localized assemblies.
• Repetitive Region Characterization: Characterize complex repetitive sequences, including tandem repeats and transposable elements, at single-base resolution.
• Insertion Analysis: Catalog and characterize insertion events genome-wide, including processed pseudogenes and long insertions (>10 kbp).
• Mechanistic Studies: Investigate associations between insertion mechanisms, short tandem duplications, transposons, and gene expression.

Methodology:

Uses minimap2 for read mapping, MAFFT for multiple sequence alignment, consensus sequence construction, a proprietary algorithm for diploid haplotype classification based on structural variants and consensus sequences, and defines target regions from mapping patterns.