# Zero-Shot 3D Drug Design by Sketching and Generating

arXiv: https://arxiv.org/abs/2209.13865 · HTML: https://mlsanity.com/paper/2209.13865/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Sep 2022.

## Built on by (most cited first, 24 of 24)

- [DecompDiff: Diffusion Models with Decomposed Priors for Structure-Based Drug Design](https://mlsanity.com/paper/2403.07902/) (2024)
- [MolCRAFT: Structure-Based Drug Design in Continuous Parameter Space](https://mlsanity.com/paper/2404.12141/) (2024)
- [3D molecule generation by denoising voxel grids](https://mlsanity.com/paper/2306.07473/) (2023)
- [A Survey of Geometric Graph Neural Networks: Data Structures, Models and Applications](https://mlsanity.com/paper/2403.00485/) (2024)
- [DecompOpt: Controllable and Decomposed Diffusion Models for Structure-based Molecular Optimization](https://mlsanity.com/paper/2403.13829/) (2024)
- [Learning Subpocket Prototypes for Generalizable Structure-based Drug Design](https://mlsanity.com/paper/2305.13997/) (2023)
- [Structure-based drug design by denoising voxel grids](https://mlsanity.com/paper/2405.03961/) (2024)
- [TAGMol: Target-Aware Gradient-guided Molecule Generation](https://mlsanity.com/paper/2406.01650/) (2024)
- [Generating 3D Binding Molecules Using Shape-Conditioned Diffusion Models with Guidance](https://mlsanity.com/paper/2502.06027/) (2025)
- [Rethinking Specificity in SBDD: Leveraging Delta Score and Energy-Guided Diffusion](https://mlsanity.com/paper/2403.12987/) (2024)
- [Unified Guidance for Geometry-Conditioned Molecular Generation](https://mlsanity.com/paper/2501.02526/) (2025)
- [Shape-conditioned 3D Molecule Generation via Equivariant Diffusion Models](https://mlsanity.com/paper/2308.11890/) (2023)
- [PharmacoNet: Accelerating Large-Scale Virtual Screening by Deep Pharmacophore Modeling](https://mlsanity.com/paper/2310.00681/) (2023)
- [3DMolFormer: A Dual-channel Framework for Structure-based Drug Discovery](https://mlsanity.com/paper/2502.05107/) (2025)
- [3D Molecular Generation via Virtual Dynamics](https://mlsanity.com/paper/2302.05847/) (2023)
- [Unified all-atom molecule generation with neural fields](https://mlsanity.com/paper/2511.15906/) (2025)
- [Geometric Deep Learning for Structure-Based Drug Design: A Survey](https://mlsanity.com/paper/2306.11768/) (2023)
- [MolChord: Structure-Sequence Alignment for Protein-Guided Drug Design](https://mlsanity.com/paper/2510.27671/) (2025)
- [CAGenMol: Condition-Aware Diffusion Language Model for Goal-Directed Molecular Generation](https://mlsanity.com/paper/2604.11483/) (2026)
- [Structure-Based Drug Design via 3D Molecular Generative Pre-training and Sampling](https://mlsanity.com/paper/2402.14315/) (2024)
- [ProtPainter: Draw or Drag Protein via Topology-guided Diffusion](https://mlsanity.com/paper/2504.14274/) (2025)
- [Delta Score: Improving the Binding Assessment of Structure-Based Drug Design Methods](https://mlsanity.com/paper/2311.12035/) (2023)
- [TransDiffSBDD: Causality-Aware Multi-Modal Structure-Based Drug Design](https://mlsanity.com/paper/2503.20913/) (2025)
- [MSCoD: An Enhanced Bayesian Updating Framework with Multi-Scale Information Bottleneck and Cooperative Attention for Structure-Based Drug Design](https://mlsanity.com/paper/2509.25225/) (2025)

## Abstract (arXiv)

Drug design is a crucial step in the drug discovery cycle. Recently, various deep learning-based methods design drugs by generating novel molecules from scratch, avoiding traversing large-scale drug libraries. However, they depend on scarce experimental data or time-consuming docking simulation, leading to overfitting issues with limited training data and slow generation speed. In this study, we propose the zero-shot drug design method DESERT (Drug dEsign by SkEtching and geneRaTing). Specifically, DESERT splits the design process into two stages: sketching and generating, and bridges them with the molecular shape. The two-stage fashion enables our method to utilize the large-scale molecular database to reduce the need for experimental data and docking simulation. Experiments show that DESERT achieves a new state-of-the-art at a fast speed.