Molecular Biological Similarity with Conditional Variational Autoencoder

Modeling the molecular biological similarity with conditional variational autoencoder. The model is trained on the ChEMBL dataset using BioBricks.

A presentation of this work is available on google drive ChemHarmony and Biosim

Development

1. install docker and nvidia-docker2
2. test nvidia-docker:
   docker run --rm --gpus all nvidia/cuda:11.7.1-devel-ubuntu20.04 nvidia-smi
3. build dockerfile
4. run dockerfile:
   docker run -p 6515:6515 -v .:/chemsim --rm --gpus all -it --name chemsim biobricks-ai/cvae

Run service

docker run -p 6515:6515 -v .:/chemsim --rm --gpus all -it --name chemsim biobricks-ai/cvae

curl -X GET "http://localhost:6515/predict?property_token=5042&inchi=InChI=1S/C9H8O4/c1-6(10)13-8-5-3-2-4-7(8)9(11)12/h2-5H,1H3,(H,11,12)"

ssh -Nf -R 12000:localhost:6515 [email protected]

curl -X GET "https://api.insilica.co/service/run/chemsim/predict?property_token=5042&inchi=InChI=1S/C9H8O4/c1-6(10)13-8-5-3-2-4-7(8)9(11)12/h2-5H,1H3,(H,11,12)"

References

• Conditional Variational Autoencoder for Molecular Biological Similarity

License

MIT