Weighing the benefits of RNA-seq error correction
Sequencing data contains two major types of errors, ones that are incorporated during library preparation and ones incorporated during sequence reading. While errors of the former are difficult to correct as they occur without clues from the base quality scores, the latter type do correlate with lower base quality scores and so it is possible to identify these ambiguous base calls and compare them to a library of high confidence k-mers from the same sequencing run. Error correction of this type has been show to improve de novo assemblies and SNP detection.
A recent report posted in bioRxiv shows error correction gives a drastic improvement in transcriptome assembly, with the author benchmarking the performance of five different correction software packages (Bless, Lighter, SGA, SEECER & BFC). The author reports that these software options vary in the aggressiveness of error correction, with SGA being the most aggressive, Lighter the least aggressive and BFC and SEECER performing w…
A recent report posted in bioRxiv shows error correction gives a drastic improvement in transcriptome assembly, with the author benchmarking the performance of five different correction software packages (Bless, Lighter, SGA, SEECER & BFC). The author reports that these software options vary in the aggressiveness of error correction, with SGA being the most aggressive, Lighter the least aggressive and BFC and SEECER performing w…