Clover

Clover performs clustering-oriented de novo assembly of Illumina sequencing reads by combining de Bruijn graph methods with a k-mer clustering approach derived from the overlap-layout-consensus concept to improve robustness to sequencing errors, particularly when using large k-mers.

Key Features:

• Clustering-oriented de novo assembly: groups k-mers during assembly to form clusters that guide contig construction.
• De Bruijn graph methodology: uses de Bruijn graphs as the core graph-based assembly framework.
• k-mer clustering (overlap-layout-consensus derived): integrates a k-mer clustering approach derived from overlap-layout-consensus concepts.
• Robustness to sequencing errors and large k-mers: mitigates Illumina sequencing errors and supports use of large k-mer sizes.
• Benchmarking metrics: reported improvements in corrected N50 and E-size metrics.
• Comparative benchmarking: evaluated against ABySS, SOAPdenovo, SPAdes, Velvet, Bambus2, CABOG, MSR-CA, and SGA.
• Datasets tested: validated on Staphylococcus aureus, Rhodobacter sphaeroides, human chromosome 14, Acinetobacter baumannii TYTH-1, and Morganella morganii KT.
• Run time performance: maintained competitive run times in comparative evaluations.

Scientific Applications:

• Microbial genome assembly: applied to bacterial genomes including Acinetobacter baumannii TYTH-1 and Morganella morganii KT.
• Bacterial and eukaryotic assembly benchmarking: used for assembly of Staphylococcus aureus, Rhodobacter sphaeroides, and human chromosome 14.
• Assembly performance evaluation: employed to improve and compare corrected N50 and E-size across assemblers.

Methodology:

Combines de Bruijn graph assembly with a k-mer clustering approach derived from overlap-layout-consensus concepts to reduce Illumina sequencing errors and enable use of large k-mers.