Investigation of biochemical diversity in a soybean lineage representing 35 years of breeding.

This paper reports an assessment of seed biochemical and metabolite variability and diversity in a series of nine soybean varieties; all lines share the same genetic lineage but represent ∼35 years of breeding (launch years 1972-2008) and differing yield potentials. These varieties, including six conventional and three glyphosate-tolerant lines, were grown concurrently at two replicated field sites in the United States during the 2011 growing season, and seeds were harvested at maturity. A compositional assessment included measurement of proximates, amino acids, fatty acids, tocopherols, isoflavones, saccharides, organic acids, and selected phytohormones. Statistical analysis included application of principal variance component analysis (PVCA) to investigate the interrelationships among compositional components from these soybean varieties and the impacts of location (environment) and pedigree on variability of these components. Results demonstrated that (i) some biochemical analytes showed trends (either increased or decreased) with launch year and/or yield, (ii) some analytes varied according to variety but showed no trend with launch year and/or yield, and (iii) almost all analytes showed extensive variation within and across sites. In summary, varietal development of high-yielding soybean, as represented in this study, has been accompanied by compositional changes but these are typically modest relative to environmental factors.

Principal variance component analysis of crop composition data: a case study on herbicide-tolerant cotton.

Compositional studies on genetically modified (GM) and non-GM crops have consistently demonstrated that their respective levels of key nutrients and antinutrients are remarkably similar and that other factors such as germplasm and environment contribute more to compositional variability than transgenic breeding. We propose that graphical and statistical approaches that can provide meaningful evaluations of the relative impact of different factors to compositional variability may offer advantages over traditional frequentist testing. A case study on the novel application of principal variance component analysis (PVCA) in a compositional assessment of herbicide-tolerant GM cotton is presented. Results of the traditional analysis of variance approach confirmed the compositional equivalence of the GM and non-GM cotton. The multivariate approach of PVCA provided further information on the impact of location and germplasm on compositional variability relative to GM.

Bayesian statistical approaches to compositional analyses of transgenic crops 2. Application and validation of informative prior distributions.

Bayesian approaches to evaluation of crop composition data allow simpler interpretations than traditional statistical significance tests. An important advantage of Bayesian approaches is that they allow formal incorporation of previously generated data through prior distributions in the analysis steps. This manuscript describes key steps to ensure meaningful and transparent selection and application of informative prior distributions. These include (i) review of previous data in the scientific literature to form the prior distributions, (ii) proper statistical model specification and documentation, (iii) graphical analyses to evaluate the fit of the statistical model to new study data, and (iv) sensitivity analyses to evaluate the robustness of results to the choice of prior distribution. The validity of the prior distribution for any crop component is critical to acceptance of Bayesian approaches to compositional analyses and would be essential for studies conducted in a regulatory setting. Selection and validation of prior distributions for three soybean isoflavones (daidzein, genistein, and glycitein) and two oligosaccharides (raffinose and stachyose) are illustrated in a comparative assessment of data obtained on GM and non-GM soybean seed harvested from replicated field sites at multiple locations in the US during the 2009 growing season.

Assessing compositional variability through graphical analysis and Bayesian statistical approaches: case studies on transgenic crops.

New transgenic (GM) crops are subjected to extensive safety assessments that include compositional comparisons with conventional counterparts as a cornerstone of the process. The influence of germplasm, location, environment, and agronomic treatments on compositional variability is, however, often obscured in these pair-wise comparisons. Furthermore, classical statistical significance testing can often provide an incomplete and over-simplified summary of highly responsive variables such as crop composition. In order to more clearly describe the influence of the numerous sources of compositional variation we present an introduction to two alternative but complementary approaches to data analysis and interpretation. These include i) exploratory data analysis (EDA) with its emphasis on visualization and graphics-based approaches and ii) Bayesian statistical methodology that provides easily interpretable and meaningful evaluations of data in terms of probability distributions. The EDA case-studies include analyses of herbicide-tolerant GM soybean and insect-protected GM maize and soybean. Bayesian approaches are presented in an analysis of herbicide-tolerant GM soybean. Advantages of these approaches over classical frequentist significance testing include the more direct interpretation of results in terms of probabilities pertaining to quantities of interest and no confusion over the application of corrections for multiple comparisons. It is concluded that a standardized framework for these methodologies could provide specific advantages through enhanced clarity of presentation and interpretation in comparative assessments of crop composition.

Introduction to Bayesian statistical approaches to compositional analyses of transgenic crops 1. Model validation and setting the stage.

Statistical comparisons of compositional data generated on genetically modified (GM) crops and their near-isogenic conventional (non-GM) counterparts typically rely on classical significance testing. This manuscript presents an introduction to Bayesian methods for compositional analysis along with recommendations for model validation. The approach is illustrated using protein and fat data from two herbicide tolerant GM soybeans (MON87708 and MON87708×MON89788) and a conventional comparator grown in the US in 2008 and 2009. Guidelines recommended by the US Food and Drug Administration (FDA) in conducting Bayesian analyses of clinical studies on medical devices were followed. This study is the first Bayesian approach to GM and non-GM compositional comparisons. The evaluation presented here supports a conclusion that a Bayesian approach to analyzing compositional data can provide meaningful and interpretable results. We further describe the importance of method validation and approaches to model checking if Bayesian approaches to compositional data analysis are to be considered viable by scientists involved in GM research and regulation.