Computational data mining and visualization of high-resolution mass spectrometry data sets for algal lipids

Lipid yields from algae are unmatched by any terrestrial feedstock, which makes their use ideal for renewable fuels and chemicals. Using algal lipids as fuel precursors has proven challenging due to the lack of complete characterization. There remain many lipid components unidentified or difficult to study. We present data on the utilization of ultra-high-resolution, fourier transform ion cyclotron resonance mass spectrometry (FT-ICR-MS) techniques and identification of novel components through novel data mining techniques. We have undertaken Kendrick mass defect analysis and Van Krevelen graphical analysis of algal lipids. In-house code has been developed to calculate Kendrick mass defect from chemical formulae and exact molecular masses for a wide range of known lipid classes found in algal oil samples and to investigate the relationship between the Kendrick mass defects and the corresponding nominal Kendrick masses. Compounds of the same class that differ from each other by the number of double bond(s), fall on a diagonal line with a well-defined slope of -6.69965. We have been able to mathematically derive the value of this slope.  We have also derived Van Krevelen diagrams (plot of atomic H/C ratio versus atomic O/C ratio) for multiple classes of algal lipids. Trends among lines in Van Krevelen plots have been used to understand structural relationships between families of lipids brought about by reactions that involve loss or gain of elements in a specific molar ratio. We explore whether Kendrick Mass defect plots and Van Krevelen diagrams, taken together, suffice as visual signatures for different algal lipid classes.