Invited review: Experimental design, data reporting, and sharing in support of animal systems modeling research1

Publication date: December 2016
Source:Journal of Dairy Science, Volume 99, Issue 12
Author(s): J.P. McNamara, M.D. Hanigan, R.R. White
The National Animal Nutrition Program "National Research Support Project 9" supports efforts in livestock nutrition, including the National Research Council's committees on the nutrient requirements of animals. Our objective was to review the status of experimentation and data reporting in animal nutrition literature and to provide suggestions for the advancement of animal nutrition research and the ongoing improvement of field-applied nutrient requirement models. Improved data reporting consistency and completeness represent a substantial opportunity to improve nutrition-related mathematical models. We reviewed a body of nutrition research; recorded common phrases used to describe diets, animals, housing, and environmental conditions; and proposed equivalent numerical data that could be reported. With the increasing availability of online supplementary material sections in journals, we developed a comprehensive checklist of data that should be included in publications. To continue to improve our research effectiveness, studies utilizing multiple research methodologies to address complex systems and measure multiple variables will be necessary. From the current body of animal nutrition literature, we identified a series of opportunities to integrate research focuses (nutrition, reproduction and genetics) to advance the development of nutrient requirement models. From our survey of current experimentation and data reporting in animal nutrition, we identified 4 key opportunities to advance animal nutrition knowledge: (1) coordinated experiments should be designed to employ multiple research methodologies; (2) systems-oriented research approaches should be encouraged and supported; (3) publication guidelines should be updated to encourage and support sharing of more complete data sets; and (4) new experiments should be more rapidly integrated into our knowledge bases, research programs and practical applications.



http://ift.tt/2g5w5Hb