Access Keys:
Skip to content (Access Key - 0)



NIRS across major food crops in the HarvestPlus Project

The purpose of establishing Near-Infrared Reflectance Spectroscopy (NIRS) was to determine pro-vitaminA components (pVAC), iron and zinc across major food crops in the HarvestPlus Project, so that selection efficiency and genetic gains for pVAC, iron and zinc in major food crops would increase. NIRS research undertaken at CIP-Lima includes eight HarvestPlus crops - wheat, rice, maize, cassava, beans, pearl millet and, as ongoing research, potato and orange-fleshed sweetpotato (OFSP). The goals are:
  • To investigate the feasibility of using NIRS for high through-put screening for provitamin A carotenoids (pVAC), iron and zinc among HarvestPlus crops.
  • To develop, install and maintain NIRS calibrations for breeders in the HarvestPlus program across different regions of the world and to give NIRS technical support to different institutions within the framework of the HarvestPlus program.
The main priorities for 2009 are to implement and to extend NIRS calibration at CGIAR-centers in order to use NIRS for pre-screening for iron and zinc (rice, wheat, pearl millet, beans and potato) and pVAC (maize, cassava and OFSP).

Maize
Several hundred maize samples were obtained from the International Center for the Improvement of Maize and Wheat (CIMMYT), scanned by NIRS and analyzed for individual carotenoids by HPLC at CIP. The samples were scanned by NIRS four times; twice as undestroyed seeds and twice as milled samples. Based on more than 200 reference values, NIRS calibrations were developed to estimate ß-cryptoxanthin, ß-carotene, zeaxanthin and lutein in undestroyed seed samples as well as in milled seed samples. The developed NIRS calibrations have been installed at CIMMYT and allow the identication of maize genotypes with elevated concentrations of pro-vitamins A (ß-cryptoxanthin and ß-carotene). Further work must be done to use the calibrations for maize samples simultaneously on both NIRS equipments at CIP and CIMMYT.

Cassava
Two hundred cassava samples obtained from International Center for Tropical Agriculture (CIAT) were scanned by NIRS and analyzed by HPLC for ß-carotene at CIP. NIRS calibrations for protein and ß-carotene were developed at CIP and installed at CIAT in Cali, Colombia. Even when the available range of ß-carotene in the analyzed sample set was very limited, low, medium and high values of ß-carotene can be distinguished by NIRS. Extension of existing NIRS calibrations for cassava with high ß-carotene samples is planned.

Beans
Several hundred bean samples were obtained from CIAT, scanned by NIRS at CIP and analyzed for iron and zinc by ICP. The samples were scanned by NIRS twice at CIP: as undestroyed seeds and as milled seed samples.
NIRS calibrations to estimate protein, iron and zinc were developed at CIP and transferred to CIAT. They are in use routinely on NIRS equipments at CIAT. In 2008, calibrations were used to estimate protein, iron and zinc concentrations in 4374 milled and 878 seed bean samples at CIAT.
Pearl Millet
In total, 170 pearl millet grain samples were obtained from the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) in India. The samples were scanned four times by NIRS at CIP: twice as undestroyed seeds and twice as milled samples. The samples were analyzed for iron and zinc by ICP. NIRS calibrations for protein, iron and zinc in milled and unmilled pearl millet samples were developed at CIP. The results clearly indicate that it is possible to use NIRS to screen for iron and zinc in pearl millet. The NIRS calibrations for grain and flour were installed at ICRISAT in 2009.

Rice
Several hundred polished and unpolished rice samples were obtained from CIAT in Colombia, and the International Rice Research Institute (IRRI) in the Philippines. The samples were scanned by NIRS at CIP four times; twice as undestroyed seeds and twice as milled samples. The samples were analyzed for iron and zinc by ICP. The calibrations for unpolished rice showed better performance than that for polished rice. The polished samples showed not only very low concentrations of iron and zinc but also not much variability in the material, which makes working with NIRS very difficult. Ninety percent of the samples have iron content between 2-4 ppm which is even difficult to distinguish by ICP. The calibrations developed at CIP were installed at CIAT for testing and pre-selection of high iron and zinc rice samples. Nevertheless these calibrations merit further improvement.
Wheat

Several hundred wheat samples were obtained from CIMMYT in Mexico. The samples were analyzed for iron and zinc by ICP. The results indicate that a pre-selection of high iron and zinc wheat samples is possible using NIRS. All developed calibrations for protein, iron and zinc were installed at CIMMYT in Mexico-City and Ciudad Obregón.