On the (Im)possibility to Reconstruct Plasmids from Whole Genome Short-Read Sequencing Data

Abstract

Plasmids are autonomous extra-chromosomal elements in bacterial cells that can carry genes that are important for bacterial survival. To benchmark algorithms for automated plasmid sequence reconstruction from short read sequencing data, we selected 42 publicly available complete bacterial genome sequences which were assembled by a combination of long- and short-read data. The selected bacterial genome sequence projects span 12 genera, containing 148 plasmids. We predicted plasmids from short-read data with four different programs (PlasmidSPAdes, Recycler, cBar and PlasmidFinder) and compared the outcome to the reference sequences. PlasmidSPAdes reconstructs plasmids based on coverage differences in the assembly graph. It reconstructed most of the reference plasmids (recall = 0.82) but approximately a quarter of the predicted plasmid contigs were false positives (precision = 0.76). PlasmidSPAdes merged 83 % of the predictions from genomes with multiple plasmids in a single bin. Recycler searches the assembly graph for sub-graphs corresponding to circular sequences and correctly predicted small plasmids but failed with long plasmids (recall = 0.12, precision = 0.30). cBar, which applies pentamer frequency composition analysis to detect plasmid-derived contigs, showed an overall recall and precision of 0.78 and 0.64. However, cBar only categorizes contigs as plasmid-derived and does not bin the different plasmids correctly within a bacterial isolate. PlasmidFinder, which searches for matches in a replicon database, had the highest precision (1.0) but was restricted by the contents of its database and the contig length obtained from de novo assembly (recall = 0.36). Surprisingly, PlasmidSPAdes and Recycler detected single isolated components corresponding to putative novel small plasmids (50 kbp) containing repeated sequences remains challenging and limits the high-throughput analysis of WGS data.