Sarcophyton glaucum is an abundant Pacific soft coral known for producing a large number of bioactive and chemically diverse secondary metabolites. Historically, S. glaucum has exhibited considerable variation in the natural products isolated, a phenomenon commonly attributed to seasonal or environmental factors. More recently, DNA sequence analysis has revealed that soft coral specimens previously classified as belonging to a single species - S. glaucum - are actually members of seven genetically-distinct clades, representing cryptic species. We hypothesize that the observed variation in the chemical profiles of these corals can largely be explained by this distinction between clades. To assess this, we collected dozens of Sarcophyton soft corals in Palau, identified them to S. glaucum clade D or F using DNA sequencing analysis, and performed gas chromatography on organic extracts of each coral specimen to obtain a chemical profile. We used both supervised and unsupervised machine learning algorithms including linear discriminant analysis, principal components analysis and hierarchical clustering to evaluate both the extent to which these genetic differences can explain the observed differences in chemical profiles and how well chemical profiles can be used to distinguish between species. We test both binning methods and discrete wavelet representations of the chemical profile for dimensionality reduction and data smoothing, and find statistical evidence that the chemical composition of the two species can be explained by distinctions between species.

Evaluating the relationship between chemical profiles and cryptic species of Sarcophyton soft corals in Palau using both supervised and unsupervised statistical learning

253rd National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.