DENTIST

DENTIST closes gaps in genome assemblies using long sequencing reads to improve assembly contiguity and base-level accuracy for downstream genomic analyses.

Key Features:

• Integration of Long Reads: Leverages long error-prone reads, such as those from PacBio continuous sequencing technologies, to increase contiguity and completeness of fragmented assemblies derived from short-read data.
• Repetitive Region Analysis: Determines repetitive regions within assembly gaps to enable reliable and unambiguous alignment of long reads to their correct loci.
• Consensus Sequence Computation: Computes consensus sequences from aligned long reads to improve the base accuracy of sequences inserted into gaps.
• Validation of Accuracy: Incorporates validation mechanisms to confirm the accuracy of the closed gaps.
• Benchmarking with Realistic Test Assemblies: Employs test assemblies with gaps positioned at the exact locations of real short-read assembly gaps and benchmarks across organisms including Drosophila, Arabidopsis, hummingbirds, and humans.
• Performance Metrics: Demonstrates higher accuracy than previous gap-closing methods while maintaining similar sensitivity in comparative studies using simulated and real PacBio long reads.

Scientific Applications:

• Genome assembly improvement: Improves contiguity and completeness of genome assemblies for downstream analyses.
• Resolution of repeat-rich genomes: Enables assembly gap resolution in complex, repeat-rich genomes of plants and animals.
• Evolutionary biology and comparative genomics: Supports cross-species analyses by providing more complete assemblies for evolutionary studies.
• Functional genomics and genetic studies: Facilitates analyses that require high base accuracy across previously gapped regions.

Methodology:

Uses long error-prone reads (e.g., PacBio) to align across gaps, determines repetitive regions within gaps to ensure unambiguous placement, computes consensus sequences from aligned reads, applies validation mechanisms for closed gaps, and benchmarks performance using test assemblies with gaps positioned at real short-read assembly gap locations evaluated with simulated and real PacBio reads.