Sharing benchmarking results for short read assembler on paired end Illumina data

[michaelbarton]

We recently wrapped up some work on benchmarking the more recent genome assemblers on a range of genomes with Illumina paired end sequence data. We compared a few of the most common metrics, one metric we included was the number of correct genes in an assembly. The overall best assembler for this metric was the JGI isolate assembly pipeline which used BBMap preprocessing steps and spades as the assembler.

The outline of this is here:

http://www.bioinformaticszen.com/post/automating-selection-of-genome-assembly-software/

A github repo for a docker image with this assembler is here:

https://github.com/bioboxes/jgi-isolate-pipeline

[pepperoni]

Would you recommend the pipeline for large genomes (100Mb) ?