Using graph convolutional neural networks to learn a representation for glycans.

Rebekka Burkholz,John Quackenbush,Daniel Bojar

doi:10.1016/j.celrep.2021.109251

Abstract

SUMMARYAs the only nonlinear and the most diverse biological sequence, glycans offer substantial challenges for computational biology. These carbohydrates participate in nearly all biological processes—from protein folding to viral cell entry—yet are still not well understood. There are few computational methods to link glycan sequences to functions, and they do not fully leverage all available information about glycans. SweetNet is a graph convolutional neural network that uses graph representation learning to facilitate a computational understanding of glycobiology. SweetNet explicitly incorporates the nonlinear nature of glycans and establishes a framework to map any glycan sequence to a representation. We show that SweetNet outperforms other computational methods in predicting glycan properties on all reported tasks. More importantly, we show that glycan representations, learned by SweetNet, are predictive of organismal phenotypic and environmental properties. Finally, we use glycan-focused machine learning to predict viral glycan binding, which can be used to discover viral receptors.

Highlights

Glycans are complex carbohydrates and are a fundamental biological sequence that is found both as isolated entities as well as covalently bound to proteins, lipids, or other molecules (Varki, 2017)
While the usage of glycowords and additional data augmentation strategies in SweetTalk partly accounted for the nonlinear nature of glycan sequences, recurrent neural networks cannot fully capture the branched or tree-like architecture that is seen in most glycans
SweetNet is a deep learning method that we developed to take advantage of the flexible graph representation structure of Graph convolutional neural networks (GCNNs)

Summary

SUMMARY

As the only nonlinear and the most diverse biological sequence, glycans offer substantial challenges for computational biology. These carbohydrates participate in most biological processes—from protein folding to viral cell entry—yet are still not well understood. SweetNet is a graph convolutional neural network that uses graph representation learning to facilitate a computational understanding of glycobiology. We show that SweetNet outperforms other computational methods in predicting glycan properties on all reported tasks. We show that glycan representations, learned by SweetNet, are predictive of organismal phenotypic and environmental properties. We use glycan-focused machine learning to predict viral glycan binding, which can be used to discover viral receptors

INTRODUCTION

RESULTS

DISCUSSION

METHOD DETAILS