Explainable graph neural networks for organic cages.

Qi Yuan,Kim E Jelfs,Filip T Szczypiński

doi:10.1039/d1dd00039j

Abstract

The development of accurate and explicable machine learning models to predict the properties of topologically complex systems is a challenge in materials science. Porous organic cages, a class of polycyclic molecular materials, have potential application in molecular separations, catalysis and encapsulation. For most applications of porous organic cages, having a permanent internal cavity in the absence of solvent, a property termed “shape persistence” is critical. Here, we report the development of Graph Neural Networks (GNNs) to predict the shape persistence of organic cages. Graph neural networks are a class of neural networks where the data, in our case that of organic cages, are represented by graphs. The performance of the GNN models was measured against a previously reported computational database of organic cages formed through a range of [4 + 6] reactions with a variety of reaction chemistries. The reported GNNs have an improved prediction accuracy and transferability compared to random forest predictions. Apart from the improvement in predictive power, we explored the explicability of the GNNs by computing the integrated gradient of the GNN input. The contribution of monomers and molecular fragments to the shape persistence of the organic cages could be quantitatively evaluated with integrated gradients. With the added explicability of the GNNs, it was possible not only to accurately predict the property of organic materials, but also to interpret the predictions of the deep learning models and provide structural insights for the discovery of future materials.

Highlights

Porous organic cages are a class of molecules with an internal cavity that is made accessible to guest molecules via at least two molecular windows (“intrinsic porosity”).[1,2] Poor packing of large organic cages in the solid-state results in accessible channels between the individual molecules (“extrinsic porosity”)
With the Graph Neural Networks (GNNs) model slightly outperforming the random forest model based upon the accuracy and precision metrics
The reason for the almost good performance of the GNN and random forest models on the All-vs-All task could originate from the dataset in this study

Summary

Introduction

Porous organic cages are a class of molecules with an internal cavity that is made accessible to guest molecules via at least two molecular windows (“intrinsic porosity”).[1,2] Poor packing of large organic cages in the solid-state results in accessible channels between the individual molecules (“extrinsic porosity”). The cavity of porous organic cages offers potential applications including encapsulation,[3] molecular separation,[4–7] and catalysis.[8]. Thanks to their molecular structure, organic cages are usually soluble in organic solvents, allowing for solution processing into thin lms or membranes both in the crystalline and amorphous solid state.[9]. The absence of three-dimensional chemical bonding allows the solid-state structures to undergo large rearrangements between polymorphs, which has been used in the creation of molecular crystals exhibiting “on/off” extrinsic porosity switching.[10]. Such exibility means that individual cage molecules are more likely to collapse and lose their

Objectives

Methods

Results

Conclusion