Advancement in sequencing technologies has been reducing sequencing costs exponentially fast. Ultra-deep sequencing is now feasible, especially for smaller genomes and clones. We expect that in the near future life scientists will sequence “as much as they want” because the sequencing cost will be a minor component of total project costs. This explosion of data will create new algorithmic challenges. Popular modern de novo assemblers are unable to take advantage of ultra-deep coverage, and the quality of assemblies starts degrading after a certain depth of coverage. SLICEMBLER is an iterative meta-assembler that solves this problem: it takes advantage of the whole dataset, and significantly improves the final quality of the assembly. SLICEMBLER partitions the input data into optimal-sized “slices” and uses a standard assembly tool (e.g., Velvet, SPAdes, IDBA, Ray) to assemble each slice individually. SLICEMBLER uses majority voting among the individual assemblies to identify long contigs that can be merged to the consensus assembly. It extracts high-quality contigs from the slice assemblies, and prevents contigs containing mis-joins and calling errors to be included in the final assembly.
SLICEMBLER has been designed and developed at the algorithm and computational biology lab. , university of California, Riverside.