Showing papers by "University of Johannesburg published in 1970"

Proteomics approach as a new way to predict tenderness as compared to the classical South African Beef Carcass Classification System

Abstract: The current South African Beef Carcass Classification System classifies carcasses using physical attributes such as age, fatness code and the conformation of the carcass as the only indicator of tenderness, implying that optimal tender meat is obtained from carcasses from animals with no permanent teeth. Research shows that if we take modern technologies (use of beta-agonists, growth hormones and electrical stimulation amongst others) into account, this is not necessarily true. In this study m. longissimus lumborum samples were collected from Nguni breed animals, snap frozen and stored at −80 oC for further use in search of protein markers that can be used to predict tenderness in a non-invasive manner. Proteins were extracted in 1 mL TES buffer, isoelectrically focused on strips (pH 5 - 7) and separated by 2D SDS PAGE, stained with coomassie brilliant blue G250) and imaged by a Chemi-doc Mp imager. The results showed differences in protein expression, e.g. some proteins are present at 1 hour post mortem (Day 0) but absent at 3 days post mortem (Day 3). Other proteins showed an increase in expression as ageing progresses. These proteins correlated with myofibril fragments and Warner Bratzler changes which were measured. There are several factors that may be responsible for these changes in differential expression during ageing such as proteolytic action (i.e. calpains and calpastatin system, cathepsins) or maybe apoptosis. The proteins in the profiles of Nguni breed animals which showed differential expression in response to ageing are still to be determined. Keywords : Nguni breed, protein markers for tenderness, two dimensional gel electrophoresis (2D SDS PAGE)

The Role of Mathematical Statistics in Improved Ore Valuation Techniques in South African Gold Mines

Abstract: Mathematical statistical techniques developed for use in ore valuation on the South African gold mines are reviewed briefly in the following categories: 1) Small random sampling theory for the three-parameter lognormal, mainly for borehole valuations; 2) Quality control techniques: a) for underground sampling based on a specially developed distribution model, b) for routine assaying based on the log-normal with variance negatively correlated with gold values; 3) Correlation, regression, and weighted moving average trend-surface techniques for routine ore reserve determinations.