Graphen Artificial Intelligence
AI tool for the future precision medicine and precision health.
Generative adversarial Neuron Network model includes two types of models. The first one is a generator, which randomly generates chemical structures. The other one is a discriminator, which classifies real and fake structures from real chemical drug database. By this way, our GAN module can simulate new drug structure. Multi-goals generative adversarial Neuron Network model will combine the affinity weights, ADME weights, toxicity weights, and druglikeness weights with target protein to generate drug candidate. No ranking selection needed.
In: PDB file or Fasta / Out: PDB file
This tool uses Seq-to-Seq way to predict drug retrosynthesis chain reaction. Based on drug structure and chemistry reaction chain information, it simulates the fastest or shortest way to synthesis small molecular drugs.
In: xyz file / Out: Retrosynthesis Reaction Chain