Fourier transform infra-red (FTIR) spectroscopy based partial least squares regression (PLSR) models were developed to assess multiple accessions of Cenchrus spp (buffel grass). The germplasm tested included accessions collected from different pastoral regions of Australia and an international set of germplasm sourced from the Australian Pasture Genebank. Drawing upon this germplasm collection, the field study described herein aimed to determine the relationships between cell wall composition, neutral detergent fiber (NDF) and indigestible NDF (iNDF) values across different strata levels and between stem and leaf tissues. Predictive models were able to identify and distinguish characteristic traits for each tissue and strata; leaf tissue possessed elevated concentrations of extractives, arabinan, galactan and ash, while stem tissue had elevated concentrations of all other variables measured. Upper strata tissue consistently had greater concentrations of acid-soluble lignin (ASL) and mannan. Of the four tissues, lower stem had the highest NDF but was also the least digestible with the highest iNDF. NDF was strongly associated with the concentration of cell wall (CW) in biomass of all tissues except for upper stem tissue, where the correlation was weaker. iNDF correlated well with higher concentrations of acid-insoluble lignin (AIL) and xylan in stem tissue, while in leaf tissue, only xylan remained closely associated with iNDF. Not only were PLSR models able to characterize the tissue types investigated, they also detected differences between the two years sampled, likely attributable to the impact of abiotic factors during the different growing seasons or the two different methods employed to clear the experimental plots.
原文链接
分享
引用
