In recent years with the huge amount of biology data in bioinformatics field, especially with the Human Genome Project, urgent needs to analyze this data to exploit optimization. Biology data characterized from other data, it directly affects the human life dramatically and significantly. In bioinformatics field there are a lot of problems need to be solved. One of the most important problem is metabolic pathway hole problem, where solving this problem helps the biologist to set the correct gene in a pathway which have a hole where the path of this pathway is unknown in some parts of it, to use these result is several useful application as gene therapy. Until now there are no enough researches to solve missing gene problem. Previous researches used BLAST as the most popular similarity tool because similar sequences usually have common descent, and therefore, similar structure and function, but these researches select some organisms from the huge amount of available organisms. In this paper we will introduce our observations of the role of organism selection and how this selection affects on the results of filling pathway hole.