Ridge Regression Estimated Linear Probability Model Predictions of N-glycosylation in Proteins with Structural and Sequence Data

TL;DR

This study develops a ridge regression model to predict N-glycosylation likelihood in human proteins using sequence and structural data, aiding experimental design without requiring prior experimental evidence.

Contribution

The paper introduces a novel ridge regression-based approach that integrates sequence and structural features to predict N-glycosylation in proteins.

Findings

Model achieves a Gini coefficient of about 74% (89%)

Predicts N-glycosylation likelihood effectively without experimental data

Incorporates amino acid distribution, structural attributes, and sequence location

Abstract

Absent experimental evidence, a robust methodology to predict the likelihood of N-glycosylation in human proteins is essential for guiding experimental work. Based on the distribution of amino acids in the neighborhood of the NxS/T sequon (N-site); the structural attributes of the N-site that include Accessible Surface Area, secondary structural elements, main-chain phi-psi, turn types; the relative location of the N-site in the primary sequence; and the nature of the glycan bound, the ridge regression estimated linear probability model is used to predict this likelihood. This model yields a Kolmogorov-Smirnov (Gini coefficient) statistic value of about 74% (89%), which is reasonable.