ABySS gave us sets of contigs for each genome, from 20 to 300 contigs per genome. We used MAUVE as a sanity check- to see that the contigs align well with known genomes of the same species, and not miss any substantial chunk of the known genomes. Also makes a pretty figure for the presentation and paper.
To link up the contigs into a single chromosome we would really need a large insert mate-pair library. A lot of work for 20 different strains. Not necessary for our approach.