Predicting the behaviour of proteins in hydrophobic interaction chromatography. 2. Using a statistical description of their surface amino acid distribution.

摘要: This paper focuses on the prediction of dimensionless retention time (DRT) proteins in hydrophobic interaction chromatography (HIC) by means mathematical models based statistical description amino acid surface distribution. Previous characterises protein as a whole. However, most it is not whole but some its specific regions that interact with environment. It seems much more natural to use local measurements characteristics surface. Therefore, characterisation distribution an property was carried out from systematic calculation average this neighbourhood placed sequentially each acids process allowed us characterise quantitatively using three main statistics: average, standard deviation and maximum. In particular, if considered hydrophobicity scale, these statistics content cluster or hotspot, well heterogeneity We tested performance DRT predictive set 15 proteins. obtained better results respect previously reported. The best model linear statistic calculated index mobilities chromatography. (measured Jack Knife MSE) 26.9% than those which does consider 19.5% imbalance (HI). addition, multivariable HI difference between experimental data smaller observed previously. fact, capacities previous decreasing MSE 8.7%. diminish number variables required, increasing, way, degrees freedom model.