Using State Machines to Improve Read Error Rates in Ion Torrent Data

In twitter and elsewhere, we are noticing renewed interest on Ion Torrent machines.

This comes on the heels of BGI’s announcement to get 37 new Proton machines.

New Announcements from Ion Torrent and Oxford Nanopore

What bioinformatics challenges are posed by IT sequencers? Ideally none, because the de Bruijn assembly algorithms and BWT-based alignment algorithms discussed in our commentaries (here and here) should work well. The issue of homolopolymers is often mentioned and readers may take a look at one interesting bioinformatics paper published recently.

Using state machines to model the Ion Torrent sequencing process and to improve read error rates

The first author David Golan created a beautiful webpage describing his method in simple language.

FlowgramFixer is an improved basecaller for the IonTorrent sequencing platform, which reduces error rates, and generates more uniquely aligned reads and more high quality reads than the default base calling algorithm implemented by IonTorrent in TorrentSuite. It is free and open source.

FlowgramFixer was developed by David Golan, Bob Harris and Paul Medvedev.

IIRC, Rayan Chikhi is also working with Paul Medvedev as a post-doc. It is one of those CS labs, which continues to publish algorithm-rich high-quality papers on NGS and bioinformatics in general.