An NIRS-based assay of chemical composition and biomass digestibility for rapid selection of Jerusalem artichoke clones

Table 1 Summary statistics for PLSR calibration models for chemical components

Parameter	N	Full spectra				CARS-optimized spectra				RF-optimized spectra
Parameter	N	RMSEC	R ² _C	RMSECV	R ² _CV	RMSEC	R ² _C	RMSECV	R ² _CV	RMSEC	R ² _C	RMSECV	R ² _CV
Soluble sugars	40	2.14	0.90	2.78	0.81	0.88	0.95	1.98	0.91	1.63	0.95	2.07	0.90
Cellulose	45	0.70	0.93	0.96	0.86	0.47	0.96	0.71	0.93	0.74	0.94	0.89	0.89
Hemicellulose	45	0.20	0.96	0.46	0.93	0.27	0.97	0.41	0.95	0.24	0.98	0.32	0.97
Lignin	45	0.48	0.88	0.87	0.77	0.49	0.92	0.71	0.85	0.42	0.95	0.56	0.92
Ash	45	0.29	0.79	0.44	0.63	0.27	0.85	0.38	0.72	0.23	0.92	0.27	0.86

N number of samples, CARS competitive adaptive reweighted sampling, RF random frog, RMSEC root mean standard error of calibration, R²_C coefficient determination of calibration, RMSECV root mean standard error of cross-validation, R²_CV coefficient determination of cross-validation

ISSN: 2731-3654