System and method for de novo drug discovery
A system and method for de novo drug discovery using machine learning algorithms. In a preferred embodiment, de novo drug discovery is performed via data enrichment and interpolation/perturbation of molecule models within the latent space, wherein molecules with certain characteristics can be generated and tested in relation to one or more targeted receptors. Filtering methods may be used to determine active novel molecules by filtering out non-active molecules and contain activity predictors to better navigate the molecule-receptor domain. The system may comprise neural networks trained to reconstruct known ligand-receptors pairs and from the reconstruction model interpolate and perturb the model such that novel and unique molecules are discovered. A second preferred embodiment trains a variational autoencoder coupled with a bioactivity model to predict molecules exhibiting a range of desired properties.
1. A system for de novo drug discovery, comprising:
a computer system comprising a memory and a processor;
a de novo discovery module, comprising a first plurality of programming instructions stored in the memory and operating on the processor, wherein the first plurality of programming instructions, when operating on the processor, causes the computer system to:
receive molecule training data comprising one or more representations of one or more chemical formulas;
enrich the molecule training data by querying data sources to find similar molecules or molecules with similar bioactivity;
use the enriched molecule training data to train an encoder of a neural network, wherein the encoder determines where each molecule in the enriched molecule training data lies in a latent space;
determine a subspace of the latent space comprising a candidate set of latent examples;
sample points within the subspace and perform interpolations, perturbations, or both on the sample points to expand the candidate set of latent examples; and
decode the candidate set of latent examples to reconstruct a candidate set of chemically valid molecules.
2. The system of claim 1 , wherein the molecule training data is formatted in standard SMILES format.
3. The system of claim 2 , wherein the SMILES format is enumerated to include non-standard information.
4. The system of claim 1 , wherein during enrichment, the de novo module interpolates between two known molecules to compensate for data disparity in the molecule training data.
5. The system of claim 1 , further comprising a bioactivity module comprising a second plurality of programming instructions stored in the memory and operating on the processor, wherein the second plurality of programming instructions, when operating on the processor, causes the computer system to rank the molecules in the candidate set of chemically valid molecules against target receptors.
6. The system of claim 1 , further comprising a reinforcement learning component that provides an additional gradient signal used to check chemical validity of decoded molecules.
7. The system of claim 6 , further comprising a reward prediction network for predicting the validity of an input graph.
8. A system for de novo drug discovery, comprising:
a computer system comprising a memory and a processor;
a de novo discovery module, comprising a first plurality of programming instructions stored in the memory and operating on the processor, wherein the first plurality of programming instructions, when operating on the processor, causes the computer system to:
receive a voxel-based representation of one or more molecules;
use the voxel-based representation of the one or more molecules to train a variational autoencoder, wherein the variational autoencoder determines where each molecule lies in a latent space; and
use the decoder portion of the variational autoencoder as a vector input to a bioactivity model, wherein the vector input comprises one or more small molecule vector representations; and
a bioactivity model comprising a second plurality of programming instructions stored in the memory and operating on the processor, wherein the second plurality of programming instructions, when operating on the processor, cause the computer system to:
train a model of voxel-based representations on a range of one or more large molecules that comprise desired molecular properties;
generate a concatenated vector comprising the one or more small molecule vector inputs and a vector representation of the one or more large molecules;
output the concatenated vector to the latent space, wherein the concatenated vector output compresses the latent space;
sample the compressed latent space, wherein the compressed latent space comprises a candidate set of latent examples; and
reconstruct the candidate set of latent examples to arrive at a candidate set of molecules that match the desired molecular properties.
9. The system of claim 8 , wherein smoothing is used to avoid underfitting.
10. The system of claim 8 , wherein constraints are applied to the latent space to avoid nonsensical atom densities.
11. The system of claim 8 , wherein the bioactivity model ranks the candidate set of molecules against target receptors.
12. A method for de novo drug discovery, comprising the steps of:
receiving molecule training data comprising one or more representations of one or more chemical formulas;
enriching the molecule training data by querying data sources to find similar molecules or molecules with similar bioactivity;
using the enriched molecule training data to train an encoder of a neural network, wherein the encoder determines where each molecule in the enriched molecule training data lies in a latent space;
determining a subspace of the latent space comprising a candidate set of latent examples;
sampling points within the subspace and perform interpolations, perturbations, or both on the sample points to expand the candidate set of latent examples; and
decoding the candidate set of latent examples to reconstruct a candidate set of chemically valid molecules.
13. The method of claim 12 , wherein the molecule training data is formatted in standard SMILES format.
14. The method of claim 13 , wherein the SMILES format is enumerated to include non-standard information.
15. The method of claim 12 , wherein during enrichment, the de novo module interpolates between two known molecules to compensate for data disparity in the molecule training data.
16. The method of claim 12 , wherein constraints are applied to the latent space to avoid nonsensical atom densities.