Graph Polish: A Novel Graph Generation Paradigm for Molecular Optimization

  title={Graph Polish: A Novel Graph Generation Paradigm for Molecular Optimization},
  author={Chaojie Ji and Yijia Zheng and Ruxin Wang and Yunpeng Cai and Hongyan Wu},
  journal={IEEE transactions on neural networks and learning systems},
Molecular optimization, which transforms a given input molecule X into another Y with desired properties, is essential in molecular drug discovery. The traditional approaches either suffer from sample-inefficient learning or ignore information that can be captured with the supervised learning of optimized molecule pairs. In this study, we present a novel molecular optimization paradigm, Graph Polish. In this paradigm, with the guidance of the source and target molecule pairs of the desired… 

A deep generative model for molecule optimization via one fragment modification

It is shown that Modof-pipe is able to retain major molecular scaffolds, allow controls over intermediate optimization steps and better constrain molecule similarities, and outperforms the state-of-the-art methods on benchmark datasets.

Molecule optimization via multi-objective evolutionary in implicit chemical space

MOMO is proposed, a multi-objective molecule optimization framework to address the challenge of efficient search for optimized molecules satisfying several properties with scarce labeled data by combining learning of chemical knowledge with Pareto-based multi- objective evolutionary search.

Smoothness Sensor: Adaptive Smoothness-Transition Graph Convolutions for Attributed Graph Clustering

A smoothness sensor for attributed graph clustering based on adaptive smoothness-transition graph convolutions is proposed, which senses the smoothness of a graph and adaptively terminates the current convolution once the Smoothness is saturated to prevent oversmoothing.



Scaffold-based molecular design with a graph generative model†

A graph generative model that targets its use in scaffold-based molecular design and can simultaneously control multiple chemical properties despite the search space constrained by fixing the substructure is proposed.

CORE: Automatic Molecule Optimization Using Copy & Refine Strategy

A new generating strategy called “Copy&Refine” (CORE), where at each step the generator first decides whether to copy an existing substructure from input X or to generate a new substructure, then the most promising substructure will be added to the new molecule.

Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation

Graph Convolutional Policy Network (GCPN) is proposed, a general graph convolutional network based model for goal-directed graph generation through reinforcement learning that can achieve 61% improvement on chemical property optimization over state-of-the-art baselines while resembling known molecules, and achieve 184% improved on the constrained property optimization task.

Junction Tree Variational Autoencoder for Molecular Graph Generation

The junction tree variational autoencoder generates molecular graphs in two phases, by first generating a tree-structured scaffold over chemical substructures, and then combining them into a molecule with a graph message passing network, which allows for incrementally expand molecules while maintaining chemical validity at every step.

Learning Multimodal Graph-to-Graph Translation for Molecular Optimization

Diverse output distributions in the model are explicitly realized by low-dimensional latent vectors that modulate the translation process and show that the model outperforms previous state-of-the-art baselines.

Generative molecular design in low data regimes

A deep learning framework for customized compound library generation is presented that aims to enrich and expand the pharmacologically relevant chemical space with drug-like molecular entities on demand to generate molecules that incorporate features of both bioactive synthetic compounds and natural products.

Automated De-Novo Design in Medicinal Chemistry: Which Types of Chemistry Does a Generative Neural Network Learn?

Different chemical spaces were used as training sets for reinforcement learning (RL) in combination with different reward functions, and 3D-shape or QSAR (quantitative structure-activity relationship) models produced significantly different proposals by sampling a broader region of the chemical space, thus supporting lead generation.

Neural Message Passing for Quantum Chemistry

Using MPNNs, state of the art results on an important molecular property prediction benchmark are demonstrated and it is believed future work should focus on datasets with larger molecules or more accurate ground truth labels.

Symmetry-adapted generation of 3d point sets for the targeted discovery of molecules

A generative neural network for 3d point sets that respects the rotational invariance of the targeted structures is introduced and demonstrated its ability to approximate the distribution of equilibrium structures using spatial metrics as well as established measures from chemoinformatics.

Deep Generative Models for 3D Linker Design

This is the first molecular generative model to incorporate 3D structural information directly in the design process, and the effectiveness and applicability of this approach on a diverse range of design problems: fragment linking, scaffold hopping, and proteolysis targeting chimera (PROTAC) design.