Hierarchical Generation of Molecular Graphs using Structural Motifs

By Wengong Jin, Regina Barzilay, Tommi S. Jaakkola. International Conference on Machine Learning, 2020

369 citations recorded by the source corpus.

Graph generation techniques are increasingly being adopted for drug discovery. Previous graph generation approaches have utilized relatively small molecular building blocks such as atoms or simple cycles, limiting their effectiveness to smaller molecules. Indeed, as we demonstrate, their performance degrades significantly for larger molecules. In this paper, we propose a new hierarchical graph encoder-decoder that employs significantly larger and more flexible graph motifs as basic building blocks. Our encoder produces a multi-resolution representation for each molecule in a fine-to-coarse fashion, from atoms to connected motifs. Each level integrates the encoding of constituents below with the graph at that level. Our autoregressive coarse-to-fine decoder adds one motif at a time, interleaving the decision of selecting a new motif with the process of resolving its attachments to the emerging molecule. We evaluate our model on multiple molecule generation tasks, including polymers, and show that our model significantly outperforms previous state-of-the-art baselines.

Subfields: Computational Theory and Mathematics, Materials Chemistry, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/dc7307f6e49f0ad2432be4eea92ae7f854c3cebb

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Materials Chemistry

In Molecular Biology

Citations in the corpus

Multi-Scale Representation Learning on Proteins (2022, methodology): “Such multi-scale representations have been previously used in molecular graph generation [Jin et al., 2020] with impressive results.”
Score-based Generative Modeling of Graphs via the System of Stochastic Differential Equations (2022, background): “…generative models include generative adversarial network (GAN) models (You et al., 2018a), recurrent neural network (RNN) models (You et al., 2018b; Popova et al., 2019), variational autoencoder (VAE) models (Jin et al., 2018; 2020), and normalizing flow models (Shi et al., 2020; Luo et al., 2021).”
Utilizing graph machine learning within drug discovery and development (2021, background): “then extended their previous work by using a hierarchical representation with various coarseness of the small molecule [103].”
Graph Representation Learning in Biomedicine and Healthcare (2021, neutral): “This is a difficult task, owing to the computational constraints of generalizability across different classes of molecules and of flexibility for a wide range of sizes 110 .”
Open Graph Benchmark: Datasets for Machine Learning on Graphs (2020, methodology): “These datasets can be used to stress-test molecule-speciﬁc methods (Yang et al., 2019; Jin et al., 2020) and transfer learning (Hu et al., 2020a).”
DiGress: Discrete Denoising diffusion for graph generation (2022, methodology): “However, these methods have not yet matched the performance of autoregressive models (Liu et al., 2018; Liao et al., 2019; Mercado et al., 2021) and motifs-based models (Jin et al., 2020; Maziarz et al., 2022), which can incorporate much more domain knowledge.”
Iterative Refinement Graph Neural Network for Antibody Sequence-Structure Co-design (2021, methodology): “The second baseline is an autoregressive graph generation model (AR-GNN) whose architecture is similar to You et al. (2018); Jin et al. (2020b) but tailored for antibodies.”