Fast Characterization of Fc-Containing Proteins by Middle-Down Mass Spectrometry Following IdeS Digestion

The rapid growth of biotherapeutics (monoclonal antibodies and antibody derivatives) demands improved techniques for their quality control and batch-to-batch study. Traditionally, the top-down and bottom-up mass spectrometric approaches may be good options, but they have their share of drawbacks. The middle-down technology has unique superiorities for fast characterization of very large proteins. In this article, we report a systematic approach to characterize Fc-containing proteins (antibody, Fc fusion protein, and antibody–drug conjugate) by middle-down mass spectrometry following IdeS proteolytic digestion. We characterized the main protein modifications including glycosylation, glycation, and other modifications with a relatively small proportion, and results were consistent with free glycan profiling analysis. Meanwhile, the O-acetylated sialic acid modification of cetuximab was reported for the first time; its potential roles in biotherapeutics need further study. The middle-down technology following IdeS proteolytic digestion could be used not only for analyzing the modifications of Fc-containing proteins at the subunit level but also for characterizing the potential modifications at early development stages.

     

ADTreesLogit model for customer churn prediction

In this paper, we propose ADTreesLogit, a model that integrates the advantage of ADTrees model and the logistic regression model, to improve the predictive accuracy and interpretability of existing churn prediction models. We show that the overall predictive accuracy of ADTreesLogit model compares favorably with that of TreeNet®, a model which won the Gold Prize in the 2003 mobile customer churn prediction modeling contest (The Duke/NCR Teradata Churn Modeling Tournament). In fact, ADTreesLogit has better predictive accuracy than TreeNet® on two important observation points.