公开(公告)号：US11615324B2

公开(公告)日：2023-03-28

申请号：US17174677

申请日：2021-02-12

申请人： Ro5 Inc.

发明人： Aurimas Pabrinkis ,  Alwin Bucher ,  Gintautas Kamuntavi{hacek over (c)}ius ,  Alvaro Prat ,  Orestis Bastas ,  {hacek over (Z)}ygimantas Jo{hacek over (c)}ys ,  Roy Tal ,  Charles Dazler Knuff

IPC分类号： G06N3/08 ,  G06N5/022 ,  G06K9/62 ,  G06F16/951

摘要： 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.

公开(公告)号：US11256995B1

公开(公告)日：2022-02-22

申请号：US17237458

申请日：2021-04-22

申请人： Ro5 Inc.

发明人： Alwin Bucher ,  Alvaro Prat ,  Orestis Bastas ,  Aurimas Pabrinkis ,  Gintautas Kamuntavi{hacek over (c)}ius ,  Mikhail Demtchenko ,  Sam Christian Macer ,  Zeyu Yang ,  Cooper Stergis Jamieson ,  {hacek over (Z)}ygimantas Jo{hacek over (c)}ys ,  Roy Tal ,  Charles Dazler Knuff

IPC分类号： G06N5/00 ,  G06N5/02 ,  G06N3/08 ,  G06K9/62 ,  G06F16/951

摘要： A system and method that predicts whether a given protein-ligand pair is active or inactive, the ground-truth protein-ligand complex crystalline-structure similarity, and an associated bioactivity value. The system and method further produce 3-D visualizations of previously unknown protein-ligand pairs that show directly the importance assigned to protein-ligand interactions, the positive/negative-ness of the saliencies, and magnitude. Furthermore, the system and method make enhancements in the art by accurately predicting protein-ligand pair bioactivity from decoupled models, removing the need for docking simulations, as well as restricting attention of the machine learning between protein and ligand atoms only.