Froth flotation is a mineral separation technique that treats large tonnages of sulfide bearing ores around the globe annually. Factors such as mineralogical composition as well as galvanic interactions between minerals surfaces and grinding media are known to control and affect the selectivity efficiency of the process. To understand these interactions, pulp chemistry information such as pH, pulp potential (Eh), dissolved oxygen (DO), oxygen demand (OD), and temperature (T) could be of great use. This work focused on unravelling potential relationships between time stamped pulp chemistry data (DO, pH and Eh) and rougher feed pyritic mineral content using statistical approaches (Pearson’s correlation coefficient criterion and mutual information). Furthermore, the potential prediction of rougher feed pyritic mineral content from DO, pH and Eh was also investigated using Gaussian process regression model. Results revealed some level of association (linear and nonlinear) between DO, pH, Eh and rougher feed pyritic mineral content. This results give room for improved process prediction based on pulp chemistry, a potential diagnostic tool for quick detection of ore variability.