Predicting minimum inhibitory concentration of antimicrobial peptides by the pseudo-amino acid composition and Gaussian kernel regression

Antimicrobial peptides (AMPs), which is a kind of short chain protein, have a strong antimicrobial ability which have antibacterial, antifungal, antiviral effect. Over the last few decades, the research of AMPs is drawing in large scholars, many of whom have engaged in the profound study on predicting AMPs activity, particularly in the AMPs classification. According to microbiology, the minimum inhibitory concentration (MIC) is a kind of antibacterial agent, its concentration is the lowest, which can inhibit the growth of the microorganism. MIC is very crucial in diagnostic lab to prove that the microbial resistance to antimicrobial agents, and to monitor the activity of new antimicrobial agents. It is generally considered as a most basic laboratory for measuring the activity of resistance on living organisms. Due to the process of biological experiments are expensive and cost plenty time, it is the highest favorable and practicable to design an efficacious computer-based MIC prediction method. In this paper, an antimicrobial peptides MIC predictor called "MIC", in which peptides sequence were formulated by incorporating five physicochemical properties into pseudo amino acid composition (PseAAC) and Gaussian kernel regression. According to the result of the MIC showed that the result of the method and experimentally result is high consistent.