BASE: A Practical de novo Assembler for Large Genomes Using Longer NGS Reads

摘要

De novo genome assembly is a fundamental problem in genomic research. When assembling relatively large genomes, time is often a very important concern, and one might have no choice but to use a more efficient assembler like SOAPdenovo2 instead of a high-quality but prohibitively slow assembler (SPAdes is a typical example). As the read length of high-throughput sequencers increases beyond lOObp, it has been expected that the quality issue of SOAPdenov2 can gradually improve. Yet SOAPdenov2, a typical de Bruijn graph based assembler, has inherent difficulty to fully utilize the advantage of longer reads (say, 150 bp and 250 bp from Illumina HiSeq and MiSeq, respectively). Other assemblers, such as the string graph assembler SGA and the multisized de Bruijn graph assembler SPAdes, though more favorable for longer reads, are very slow and less popular. It is still up to the challenge how to better utilize longer reads to develop a fast-and-accurate assembler.