My experience with the AllPaths-LG genome assembler has been great. Anecdotal evidence as well as some rigorous studies both seem to corroborate this. Earlier this week, I came across a link to this post on the AllPaths-LG blog.
The ALLPATHS-LG fragment read error correction modules are now available for stand alone use outside the full assembly pipeline. You can now easily error correct your own data using our method without running an assembly.
I’m glad I saw this when I did. I recently read over Steven Salzberg et al.‘s GAGE evaluation, and remembered the paper mentioning that for many (most?) assemblers, using AllPaths-LG’s error correction module as opposed to QUAKE gave better results. I’m definitely interested to try this out.