The evolutionary root of flowering plants.

Correct rooting of the angiosperm radiation is both challenging and necessary for understanding the origins and evolution of physiological and phenotypic traits in flowering plants. The problem is known to be difficult due to the large genetic distance separating flowering plants from other seed plants and the sparse taxon sampling among basal angiosperms. Here, we provide further evidence for concern over substitution model misspecification in analyses of chloroplast DNA sequences. We show that support for Amborella as the sole representative of the most basal angiosperm lineage is founded on sequence site patterns poorly described by time-reversible substitution models. Improving the fit between sequence data and substitution model identifies Trithuria, Nymphaeaceae, and Amborella as surviving relatives of the most basal lineage of flowering plants. This finding indicates that aquatic and herbaceous species dominate the earliest extant lineage of flowering plants. [; ; ; ; ; .].

Systematic error in seed plant phylogenomics.

Resolving the closest relatives of Gnetales has been an enigmatic problem in seed plant phylogeny. The problem is known to be difficult because of the extent of divergence between this diverse group of gymnosperms and their closest phylogenetic relatives. Here, we investigate the evolutionary properties of conifer chloroplast DNA sequences. To improve taxon sampling of Cupressophyta (non-Pinaceae conifers), we report sequences from three new chloroplast (cp) genomes of Southern Hemisphere conifers. We have applied a site pattern sorting criterion to study compositional heterogeneity, heterotachy, and the fit of conifer chloroplast genome sequences to a general time reversible + G substitution model. We show that non-time reversible properties of aligned sequence positions in the chloroplast genomes of Gnetales mislead phylogenetic reconstruction of these seed plants. When 2,250 of the most varied sites in our concatenated alignment are excluded, phylogenetic analyses favor a close evolutionary relationship between the Gnetales and Pinaceae-the Gnepine hypothesis. Our analytical protocol provides a useful approach for evaluating the robustness of phylogenomic inferences. Our findings highlight the importance of goodness of fit between substitution model and data for understanding seed plant phylogeny.

Whole genome sequencing of enriched chloroplast DNA using the Illumina GAII platform.

BACKGROUND: Complete chloroplast genome sequences provide a valuable source of molecular markers for studies in molecular ecology and evolution of plants. To obtain complete genome sequences, recent studies have made use of the polymerase chain reaction to amplify overlapping fragments from conserved gene loci. However, this approach is time consuming and can be more difficult to implement where gene organisation differs among plants. An alternative approach is to first isolate chloroplasts and then use the capacity of high-throughput sequencing to obtain complete genome sequences. We report our findings from studies of the latter approach, which used a simple chloroplast isolation procedure, multiply-primed rolling circle amplification of chloroplast DNA, Illumina Genome Analyzer II sequencing, and de novo assembly of paired-end sequence reads. RESULTS: A modified rapid chloroplast isolation protocol was used to obtain plant DNA that was enriched for chloroplast DNA, but nevertheless contained nuclear and mitochondrial DNA. Multiply-primed rolling circle amplification of this mixed template produced sufficient quantities of chloroplast DNA, even when the amount of starting material was small, and improved the template quality for Illumina Genome Analyzer II (hereafter Illumina GAII) sequencing. We demonstrate, using independent samples of karaka (Corynocarpus laevigatus), that there is high fidelity in the sequence obtained from this template. Although less than 20% of our sequenced reads could be mapped to chloroplast genome, it was relatively easy to assemble complete chloroplast genome sequences from the mixture of nuclear, mitochondrial and chloroplast reads. CONCLUSIONS: We report successful whole genome sequencing of chloroplast DNA from karaka, obtained efficiently and with high fidelity.