Prediction of protein retention times in hydrophobic interaction chromatography by robust statistical characterization of their atomic‐level surface properties

Abstract

The correlation between the dimensionless retention times (DRT) of proteins in hydrophobic interaction chromatography (HIC) and their surface properties were investigated. A ternary atomic-level hydrophobicity scale was used to calculate the distribution of local average hydrophobicity across the proteins surfaces. These distributions were characterized by robust descriptive statistics to reduce their sensitivity to small changes in the three-dimensional structure. The applicability of these statistics for the prediction of protein retention behaviour was looked into. A linear combination of robust statistics describing the central tendency, heterogeneity and frequency of highly hydrophobic clusters was found to have a good predictive capability (R2 = 0.78), when combined a factor to account for protein size differences. The achieved error of prediction was 35% lower than for a similar model based on a description of the protein surface on an amino acid level. This indicates that a robust and mathematically simple model based on an atomic description of the protein surface can be used for the prediction of the retention behaviour of conformationally stable globular proteins with a well determined 3D structure in HIC. © 2016 American Institute of Chemical Engineers Biotechnol. Prog., 32:372-381, 2016.

Predicting Antibody Developability Profiles Through Early Stage Discovery Screening

“…All DLS studies were performed at 25°C on the undiluted samples in glass bottom 96-well plates using a DynaPro Plate Reader II (Wyatt, Santa Barbara, CA). Twenty (20) acquisitions (5 seconds each) per well were averaged and the hydrodynamic radius (Rh), % polydispersity, and % mass was modeled using Dynamics version 7.1.9.3 (Wyatt Technology) to assess aggregation.…”

“…Machine learning methods (e.g., random forest) to predict the hydrophobic chromatography (HIC) retention time of a given antibody sequence were applied to develop predictive models. 19,20 Progress is being made toward establishing correlations between biophysical assays and computationally predictive behavior for downstream and manufacturing endpoints using data gathered for a large number of antibody molecules. In 2017, Jain et al reported the production and characterization using a dozen biophysical property assays of a panel of 137 monoclonal antibodies currently in advanced clinical stages.…”

Models for Antibody Behavior in Hydrophobic Interaction Chromatography and in Self-Association

“…Reports indicate that mAbs with greater retention in HIC studies may also exhibit greater F o r P e e r R e v i e w context of sequence 8 and structure 25 indicate that low hydrophobicity correlates with reduced retention time. Recent work attempts to predict solvent accessible surface area (SASA), and associated non-polar area, from amino acid sequence, and finds reasonable performance of a binary classifier for lower and higher retention times in HIC 26 .…”

SSH: A Tool for Predicting Hydrophobic Interaction of Monoclonal Antibodies Using Sequences

“…Currently, there are also quite a few existing methods for predicting the hydrophobicity of proteins including mAbs [21][22][23]. These methods are mostly based on threedimensional structures of proteins.…”