# A Two-Step Graph Convolutional Decoder for Molecule Generation

arXiv: https://arxiv.org/abs/1906.03412 · HTML: https://mlsanity.com/paper/1906.03412/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Jun 2019.

## Built on by (most cited first, 25 of 25)

- [Equivariant Diffusion for Molecule Generation in 3D](https://mlsanity.com/paper/2203.17003/) (2022)
- [SPECTRE: Spectral Conditioning Helps to Overcome the Expressivity Limits of One-shot Graph Generators](https://mlsanity.com/paper/2204.01613/) (2022)
- [GraphEBM: Molecular Graph Generation with Energy-Based Models](https://mlsanity.com/paper/2102.00546/) (2021)
- [A Survey on Deep Graph Generation: Methods and Applications](https://mlsanity.com/paper/2203.06714/) (2022)
- [Top-N: Equivariant set and graph generation without exchangeability](https://mlsanity.com/paper/2110.02096/) (2021)
- [Graph Neural Networks for Molecules](https://mlsanity.com/paper/2209.05582/) (2022)
- [Brain Multigraph Prediction using Topology-Aware Adversarial Graph Neural Network](https://mlsanity.com/paper/2105.02565/) (2021)
- [MAGNet: Motif-Agnostic Generation of Molecules from Shapes](https://mlsanity.com/paper/2305.19303/) (2023)
- [Any2Graph: Deep End-To-End Supervised Graph Prediction With An Optimal Transport Loss](https://mlsanity.com/paper/2402.12269/) (2024)
- [The power of motifs as inductive bias for learning molecular distributions](https://mlsanity.com/paper/2306.17246/) (2023)
- [GraphGANFed: A Federated Generative Framework for Graph-Structured Molecules Towards Efficient Drug Discovery](https://mlsanity.com/paper/2304.05498/) (2023)
- [Edge but not Least: Cross-View Graph Pooling](https://mlsanity.com/paper/2109.11796/) (2021)
- [Learning to Untangle Genome Assembly with Graph Convolutional Networks](https://mlsanity.com/paper/2206.00668/) (2022)
- [Geometric learning of the conformational dynamics of molecules using dynamic graph neural networks](https://mlsanity.com/paper/2106.13277/) (2021)
- [Graph Context Encoder: Graph Feature Inpainting for Graph Generation and Self-supervised Pretraining](https://mlsanity.com/paper/2106.10124/) (2021)
- [An In-depth Summary of Recent Artificial Intelligence Applications in Drug Design](https://mlsanity.com/paper/2110.05478/) (2021)
- [End-to-End Graph Flattening Method for Large Language Models](https://mlsanity.com/paper/2409.14880/) (2024)
- [Molecular modeling with machine-learned universal potential functions](https://mlsanity.com/paper/2103.04162/) (2021)
- [Bayesian Deep Learning for Graphs](https://mlsanity.com/paper/2202.12348/) (2022)
- [Graph-level representations using ensemble-based readout functions](https://mlsanity.com/paper/2303.02023/) (2023)
- [Anchor Space Optimal Transport as a Fast Solution to Multiple Optimal Transport Problems](https://mlsanity.com/paper/2310.16123/) (2023)
- [Heterogeneous Graph Generation: A Hierarchical Approach using Node Feature Pooling](https://mlsanity.com/paper/2410.11972/) (2024)
- [Auto-encoding Molecules: Graph-Matching Capabilities Matter](https://mlsanity.com/paper/2503.00426/) (2025)
- [GIN-Graph: A Generative Interpretation Network for Model-Level Explanation of Graph Neural Networks](https://mlsanity.com/paper/2503.06352/) (2025)
- [Conditional GraphGANFed: Optimizing Graph-Structured Molecule Generation in Federated Generative Adversarial Networks](https://mlsanity.com/paper/2608.24610/) (2026)

## Abstract (arXiv)

We propose a simple auto-encoder framework for molecule generation. The molecular graph is first encoded into a continuous latent representation z, which is then decoded back to a molecule. The encoding process is easy, but the decoding process remains challenging. In this work, we introduce a simple two-step decoding process. In a first step, a fully connected neural network uses the latent vector z to produce a molecular formula, for example CO2 (one carbon and two oxygen atoms). In a second step, a graph convolutional neural network uses the same latent vector z to place bonds between the atoms that were produced in the first step (for example a double bond will be placed between the carbon and each of the oxygens). This two-step process, in which a bag of atoms is first generated, and then assembled, provides a simple framework that allows us to develop an efficient molecule auto-encoder. Numerical experiments on basic tasks such as novelty, uniqueness, validity and optimized chemical property for the 250k ZINC molecules demonstrate the performances of the proposed system. Particularly, we achieve the highest reconstruction rate of 90.5%, improving the previous rate of 76.7%. We also report the best property improvement results when optimization is constrained by the molecular distance between the original and generated molecules.