Evaluation of corn endosperm properties and development of near infrared reflectance spectroscopy (NIRS) calibrations for screening germplasm on starch digestibility in large scale corn breeding programs | | Posted on:2006-03-26 | Degree:Ph.D | Type:Dissertation | | University:The University of Wisconsin - Madison | Candidate:Ngonyamo-Majee, Dainah | Full Text:PDF | | GTID:1451390008462550 | Subject:Agriculture | | Abstract/Summary: | PDF Full Text Request | | Corn (Zea mays L.) starch forms the main source of energy in many parts of the world, contributing 50% and 75% of the energy values of silage and grain, respectively. Hence, evaluating kernel endosperm properties influencing starch availability and developing near infrared reflectance spectroscopy calibrations (NIRS) would allow reduced analytical costs and faster data turn around in future large-scale hybrid evaluation trials.; Several experiments were conducted to evaluate the effects of single gene mutations, different inbred lines and hybrids harvested at different maturities. Near-infrared reflectance spectroscopy (NIRS) calibrations for corn vitreousness (%V), density (D) and Stenvert hardness measurements of time to collect ground sample to a set receptacle height (T) and total column height (CH) were developed as indirect predictors of degradability. The NIRS calibrations for %V, D and Stenvert T and CH had high R2-values (0.90; 0.92; 0.85 and 0.85 respectively). These results show that NIRS calibrations on endosperm properties can be used to indirectly screen corn genetic material for degradability in large-scale breeding trials. A direct MRS calibration on degradability was developed using data generated from experiments involving US, South American and Zimbabwean corn germplasm. Calibration statistics from these data sets showed high R2 values, low standard error of calibration (SEC) and low standard error of cross validation (SECV), which are all indicators of a good NIRS calibration. Laboratory vs. NIRS predicted values from the combined data set (547 samples) showed strong correlations (r = 0.90; 0.92; and 0.95), for A-fraction, RDMD and TDMD, respectively. Hence, to ensure greater labor efficiency, reduced analytical costs, and faster data turn around, we recommend the use of these NIRS predictions for ranking corn germplasm in breeding trials for starch properties and ruminant in situ or in vitro measurements. | | Keywords/Search Tags: | NIRS, Corn, Starch, Endosperm properties, Reflectance spectroscopy, Germplasm, Breeding, Calibrations | PDF Full Text Request | Related items |
| |
|