Cerulean

Cerulean performs hybrid genome assembly by combining short high-fidelity reads and long Pacific Biosciences reads to resolve repetitive regions and improve assembly accuracy from high-throughput sequencing data.

Key Features:

• Short-read assembly: Assembles accurate short-read data into contigs and an assembly graph to provide a reliable backbone for the genome.
• Long-read mapping: Maps long reads (e.g., Pacific Biosciences) onto the short-read assembly graph instead of performing direct long-read error correction.
• Assembly graph usage: Uses the short-read assembly graph as the reference structure for incorporating long-read information.
• Repeat resolution and gap filling: Leverages long-read mappings to resolve repetitive regions and fill gaps that short reads cannot span.
• Reduced computational demands: Avoids computationally intensive direct error correction of long reads to lower resource usage.
• Improved accuracy: Combines high fidelity of short reads with the spanning capability of long reads to enhance assembly correctness.
• Scalability: Designed to handle large sequencing datasets for a range of genome sizes.

Scientific Applications:

• Repetitive region resolution: Resolves complex and highly repetitive genomic regions by using long reads to bridge repeats in the assembly graph.
• Genome assembly from high-throughput sequencing: Produces improved assemblies from combined short- and long-read high-throughput sequencing datasets.
• Bacterial genome assembly: Enables assembly of small bacterial genomes using hybrid data while managing large datasets.

Methodology:

Assemble short reads into contigs and an assembly graph, map long reads onto that assembly graph instead of correcting long reads, and use long-read mappings to resolve repeats and fill gaps.