Development of a chemically intuitive filter for chemical graph convolutional network

Abstract

This work proposes the use of I-N - A (I-N: identity matrix; A: adjacency matrix), instead of I-N + A, the normalized form of which has intensively been used for the construction of graph convolutional networks (GCNs), in deep-learning chemistry. The performance of the GCN model with D-1/2(I-N - A)D-1/2 in its convolution step is at least on a par with the vanilla GCN that uses (D) over tilde (-1/2)(I-N + A)(D) over tilde (-1/2) ((D) over tilde: degree matrix of I-N + A) in various chemistry datasets, such as FreeSolv, ESOL, lipophilicity, and blood-brain barrier penetration datasets. It could be seen that the use of I-N - A might be more chemically intuitive than the use of I-N + A, potentially embracing the information on bond properties, such as dipole moment, and functional groups in a molecule. This work suggests unavoidable necessity of tackling molecular-representation problems in deep-learning chemistry from unprecedented angles of view for advanced development and construction of chemically intuitive deep-learning models.