Fourier‐transform mid‐infrared spectroscopy for in vitro protein digestibility measurement of pulse crops

Abstract

Abstract The standard method of estimating in vitro protein digestibility, the protein digestibility corrected amino acid score (PDCAAS) assay, does not support the expected workflow of a pulse breeding program. This is mainly due to its low‐throughput design and long processing time (∼16–24 h) per sample. Fourier‐transform mid‐infrared (FT‐MIR) spectroscopy has been developed as a high‐throughput phenotyping tool to estimate protein digestibility in pulses. The mid‐infrared region representing the amide I band (1756.81–1586.27 cm −1 ) was utilized to perform chemometric modeling with partial least squares regression (PLSR) to estimate in vitro protein digestibility in dry pea ( Pisum sativum L.), lentil ( Lens culinaris Medik.), and chickpea ( Cicer arietinum L.) flours. The root mean square error of predictions of the developed PLSR models for dry pea, lentil, and chickpea were 0.00039, 0.00024, and 0.00017, respectively. Phenotyping with the FT‐MIR approach is more rapid than with the PDCAAS assay and estimates protein digestibility from the flour of a single seed in a shorter time (∼1–2 min). The FT‐MIR approach is robust as the spectroscopic data are consistent and chemically fingerprint this nutritional trait. Accordingly, FT‐MIR can resolve the phenotypic bottleneck of pulse breeding related to in vitro protein digestibility measurements by enabling a high‐throughput phenotyping workflow.

Code and data availability

The article describes FT-MIR spectral datasets and PLSR models for pulse protein digestibility, but no public repository, data deposit, or author code URL is provided. The only availability statement is a generic Supporting Information note without a link, and all URLs in the text are references, publisher access pages

No evidence-backed public reproduction asset is currently recorded.