Friday, 11 May 2012

Flash Report: Life Technologies official reply to Loman's paper

Life Technologies has posted a reply to the "Performance comparison of benchtop high-throughput sequencing platforms" manuscript published in April on Nature Biotechnology.
Here is what they did:

How did we address some of the challenges?

• We obtained a sample of the EHEC E. coli isolate, the week after the paper was released
• We sequenced this sample using up-to-date Ion sequencing kits and software
• We directly compared the data from Loman et al July 2011 to Ion publicly released data from May 2012*
• We aligned the reads using the genome sequenced in Loman et al to measure accuracy
• We performed de novo assembly to measure how well the PGM performs for this application

Using the most recent chemistries, algorithms and the 318 Chip, the Ion Torrent PGM error rates are definitely lower (the majority of bases are Q30 or better), although in this area the MiSeq seems to have an edge on the competition, particularly in the homopolymer stretches.
It is however remarkable that in 9 months (July 2011 - April 2012) the indel and homopolymer accuracy has improved ~450% on the Ion Torrent platform.

No comments: