Premium
Chemical variance of dry beans determined from UV spectral fingerprints
Author(s) -
Harnly James M,
Luthria Dave,
PastorCorrales Marcial S
Publication year - 2009
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.23.1_supplement.923.1
Subject(s) - principal component analysis , analysis of variance , statistics , mathematics , variance (accounting) , coefficient of variation , yield (engineering) , one way analysis of variance , statistical analysis , analytical chemistry (journal) , chemistry , chromatography , materials science , metallurgy , business , accounting
The chemical variance of dry beans arising from three locations (Maryland, Michigan, and Nebraska), 9 varieties, 8 row samples, and triplicate determinations was determined from UV spectral fingerprints analyzed by analysis of variance‐principal components analysis (ANOVA‐PCA). Samples were extracted with 60% aqueous methanol, diluted as appropriate, and spectra (200‐400 nm) acquired by UV spectrophotometry. The data were analyzed using nested one‐way ANOVA where classic single variables were replaced by the sum of results for all wavelengths (200 variables). ANOVA‐PCA allowed pretreatment of the data matrix (648 samples x 200 wavelengths) to yield separate sub‐matrices isolating each experimental factor (location, variety, and row). Visual and statistical interpretation was easily accomplished. The results showed that, for 25 plants per row, 0.3% of the total variance came from location, 10.2% from variety, 70.8% from plant‐to‐plant variation, and 0.9% from analytical uncertainty. This research was funded by USDA and the Office of Dietary Supplements at the National Institutes of Health.