An algorithm for automated closure during assembly

Abstract

Background: Finishing is the process of improving the quality and utility of draft genome sequences generated by shotgun sequencing and computational assembly. Finishing can involve targeted sequencing. Finishing reads may be incorporated by manual or automated means. One automated method uses targeted addition by local re-assembly of gap regions. An obvious alternative uses de novo assembly of all the reads.

Results: A procedure called the bounding read algorithm was developed for assembly of shotgun reads plus finishing reads and their constraints, targeting repeat regions. The algorithm was implemented within the Celera Assembler software and its pyrosequencing-specific variant, CABOG. The implementation was tested on Sanger and pyrosequencing data from six genomes. The bounding read assemblies were compared to assemblies from two other methods on the same data. The algorithm generates improved assemblies of repeat regions, closing and tiling some gaps while degrading none.

Conclusions: The algorithm is useful for small-genome automated finishing projects. Our implementation is available as open-source from http://wgs-assembler.sourceforge.net under the GNU Public License.

Figure 1

Use of finishing reads. Two algorithms for assembling shotgun reads and finishing reads. The control treats both read types equally. The bounded algorithm attempts to assemble finishing reads consistently with their bounding constraints. For each algorithm, the figure shows its construction of a scaffold from contigs (rectangles) with 2X in shotgun reads (black lines). Each finishing read (colored line) has a corresponding pair of PCR primer sites (arrows of same color). External to the scaffold is a unitig (grey area) deemed repetitive due to high coverage. (a) A mate pair constraint (curve) localizes one read and the unitig to this gap. Nevertheless, the control algorithm cannot tile this gap with reads. The bounded algorithm localizes two finishing reads by their primer sites. The bounded algorithm does tile the gap with reads, enabling a more accurate consensus sequence. (b) The control cannot localize the unitig or any reads to this gap. It does not close the gap. The bounded algorithm localizes the unitig by finishing reads and their primer sites. It tiles the gap with finishing reads from the unitig. (c) Both algorithms assemble finishing reads from a gap that is not a genomic repeat. In our data sets, most finishing reads fit gaps of this type.