Fuzzy Hybrid least-Squares Regression Approach to Estimating the amount of Extra Cellular Recombinant Protein A from Escherichia coli BL21

Introduction: Immune Protein A is a component with a vast spectrum of biochemical, biological and medical usages. The coding gene of this protein was extracted from Staphylococcus aureus and was cloned and expressed in Escherichia coli bacteria. Suitable statistical methods are utilized to optimize expression conditions  for evaluating experiment accuracy , guarantee the accuracy of subsequent experiments, reduce cost , prevent trial and error method, and obtain the highest production level.   Materials & Methods: Normal statistical regression method is based on the assumption of accuracy of variables and their observations, and finally the relationship among the variables are precisely specified .In normal modeling ,such as estimation of protein level inaccurate observations and vague relationships may be encountered; therefore,utilization of regression methods capable of explaining the vague structure of protein level and providing the models attunes to reality is necessary. In this article, hybrid fuzzy regression method with the least square, based on fuzzy-set theory was utilized.   Findings: According to the results, Protein A production level was estimated at 90 % level .On the other hand, since the method utilized in this article was hybrid fuzzy linear regression method with the least square errors, we can conclude that if all the data utilized in this article are crisp numbers,the hybrid fuzzy regression will produce results similar to normal regression.   Discussion & Conclusions: One of the advantages of hybrid fuzzy regression is higher level of certainty than point estimation. In this study, the standard deviation of estimated data by fuzzy hybrid regression method was lower than conventional regression method.Therefore, it can be concluded that fuzzy hybrid regression method is an optimal method for estimating the production of this type of protein.