Proteins play different roles in the body, and the task of understanding proteins function is very important in biological processes. Proteins are composed of linear sequences of smaller molecules called amino acids that accomplish most of the functions of the living cell. There are twenty different amino acids whose chemical and physical properties are varied. Any protein that is produced has two ends: the C-terminal and the N-terminal. Many sources of information are used to understand protein function. The goals in cell biology are to identify the subcellular locations of proteins. The understanding of subcellular location of proteins is very important to understand their function. So knowing the location of proteins within the cell is an important step to understand the function of protein as well as the role in biological processes. If Proteins located in the correct subcellular locations only then protein perform their appropriate functions. Biochemical experiments are required to determine the subcellular localization of a protein, but experiments are time consuming and high effort requires. This is motivation behind proposed technique. So it needs to develop computational system to predict protein subcellular localization automatically and accurately.
Three types
…show more content…
The traditional protein subcellular localization systems can only handle single label problems. However, protein may exist in more than one subcellular location or they move from one cell to another cell. So that it needs to assigns possibly more than one label per location as the result of classification. The difference between single-label classification and multi-label classification is that the aim of multi-label classification is looking for a group of labels which is related to the data instead of a single