# Valid Property-Enhanced Contrastive Learning for Targeted Optimization & Resampling for Novel Drug Design

arXiv: https://arxiv.org/abs/2509.00684 · HTML: https://mlsanity.com/paper/2509.00684/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 86 references that are in the map.
- First posted on arXiv 31 Aug 2025.

## Builds on (most cited first, 23 of 23)

- [Automatic chemical design using a data-driven continuous representation of molecules](https://mlsanity.com/paper/1610.02415/) (2016)
- [Junction Tree Variational Autoencoder for Molecular Graph Generation](https://mlsanity.com/paper/1802.04364/) (2018)
- [Prototypical Contrastive Learning of Unsupervised Representations](https://mlsanity.com/paper/2005.04966/) (2020)
- [GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generation](https://mlsanity.com/paper/2203.02923/) (2022)
- [Molecular De Novo Design through Deep Reinforcement Learning](https://mlsanity.com/paper/1704.07555/) (2017)
- [Contrastive and Non-Contrastive Self-Supervised Learning Recover Global and Local Spectral Embedding Methods](https://mlsanity.com/paper/2205.11508/) (2022)
- [Molecular generative model based on conditional variational autoencoder for de novo molecular design](https://mlsanity.com/paper/1806.05805/) (2018)
- [Conditional molecular design with deep generative models](https://mlsanity.com/paper/1805.00108/) (2018)
- [DrugAgent: Automating AI-aided Drug Discovery Programming through LLM Multi-Agent Collaboration](https://mlsanity.com/paper/2411.15692/) (2024)
- [Molecular Generation with Recurrent Neural Networks (RNNs)](https://mlsanity.com/paper/1705.04612/) (2017)
- [Text-Guided Molecule Generation with Diffusion Language Model](https://mlsanity.com/paper/2402.13040/) (2024)
- [Artificial Intelligence in Drug Discovery: Applications and Techniques](https://mlsanity.com/paper/2106.05386/) (2021)
- [RAG-Enhanced Collaborative LLM Agents for Drug Discovery](https://mlsanity.com/paper/2502.17506/) (2025)
- [ACEGEN: Reinforcement learning of generative chemical agents for drug discovery](https://mlsanity.com/paper/2405.04657/) (2024)
- [Solving Inverse Problems in Protein Space Using Diffusion-Based Priors](https://mlsanity.com/paper/2406.04239/) (2024)
- [Understanding Self-Supervised Learning via Gaussian Mixture Models](https://mlsanity.com/paper/2411.03517/) (2024)
- [Synergistic Benefits of Joint Molecule Generation and Property Prediction](https://mlsanity.com/paper/2504.16559/) (2025)
- [Phenotypic Profile-Informed Generation of Drug-Like Molecules via Dual-Channel Variational Autoencoders](https://mlsanity.com/paper/2506.02051/) (2025)
- [Zero-Shot Learning with Subsequence Reordering Pretraining for Compound-Protein Interaction](https://mlsanity.com/paper/2507.20925/) (2025)
- [Surprisal Driven k-NN for Robust and Interpretable Nonparametric Learning](https://mlsanity.com/paper/2311.10246/) (2023)
- [Revealing the Relationship Between Publication Bias and Chemical Reactivity with Contrastive Learning](https://mlsanity.com/paper/2402.16882/) (2024)
- [Generative molecule evolution using 3D pharmacophore for efficient Structure-Based Drug Design](https://mlsanity.com/paper/2507.20130/) (2025)
- [Adaptive Multimodal Protein Plug-and-Play with Diffusion-Based Priors](https://mlsanity.com/paper/2507.21260/) (2025)

## Abstract (arXiv)

Efficiently steering generative models toward pharmacologically relevant regions of chemical space remains a major obstacle in molecular drug discovery under low-data regimes. We present VECTOR+: Valid-property-Enhanced Contrastive Learning for Targeted Optimization and Resampling, a framework that couples property-guided representation learning with controllable molecule generation. VECTOR+ applies to both regression and classification tasks and enables interpretable, data-efficient exploration of functional chemical space. We evaluate on two datasets: a curated PD-L1 inhibitor set (296 compounds with experimental IC50 values) and a receptor kinase inhibitor set (2,056 molecules by binding mode). Despite limited training data, VECTOR+ generates novel, synthetically tractable candidates. Against PD-L1 (PDB 5J89), 100 of 8,374 generated molecules surpass a docking threshold of -15.0 kcal/mol, with the best scoring -17.6 kcal/mol compared to the top reference inhibitor (-15.4 kcal/mol). The best-performing molecules retain the conserved biphenyl pharmacophore while introducing novel motifs. Molecular dynamics (250 ns) confirm binding stability (ligand RMSD < 2.5 angstroms). VECTOR+ generalizes to kinase inhibitors, producing compounds with stronger docking scores than established drugs such as brigatinib and sorafenib. Benchmarking against JT-VAE and MolGPT across docking, novelty, uniqueness, and Tanimoto similarity highlights the superior performance of our method. These results position our work as a robust, extensible approach for property-conditioned molecular design in low-data settings, bridging contrastive learning and generative modeling for reproducible, AI-accelerated discovery.