ALCHIMIA: Generated Molecules for CB2R and S1R Targets via RL-Guided Genetic Algorithm

A dataset of molecules generated by the ALCHIMIA framework for computational drug discovery. The framework uses a vocabulary of 33 medicinal chemistry-inspired transformations to optimize synthetic accessibility, drug-likeness, and binding affinity for two human protein targets. The dataset, 1.7 MB in size, was created by Domenico Alberga and last updated on April 16, 2026.

Use Cases

• Benchmarking generative molecular design models based on the described QED and SA scores.
• Exploring chemical lineages for the human Cannabinoid Receptor 2 (CB2R) based on the described docking-guided optimization.
• Exploring chemical lineages for the human Sigma nonopioid intracellular Receptor 1 (S1R) based on the described docking-guided optimization.
• Training or validating reinforcement learning policies for molecular transformation sequences.
• Studying molecules generated under scaffold-constrained or dual-modulator design scenarios as described.

Strengths

• Dataset is associated with a specific, interpretable framework (ALCHIMIA) combining reinforcement learning and a genetic algorithm.
• Molecules were generated using a defined vocabulary of 33 medicinal chemistry-inspired transformations.
• Framework performance was evaluated against specific pharmacological targets (CB2R and S1R) and design scenarios.
• The dataset is freely available via a public GitHub repository.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare

Collection Method

Generated by the ALCHIMIA computational framework.

Freshness

Last updated 2026-04-16 07:29:42; freshness should be verified.