Optimizing molecule toxicity by replacing target fragments with bioisosteres

A system comprising a hardware processor and a non-transitory computer-readable storage medium storing executable instructions that, when executed by the hardware processor, cause the hardware processor to perform steps comprising:

• obtaining data identifying:
• 1) an input molecule, and 2) a set of features characterizing the input molecule;
• for each dose value in a sequence of dose values, generating a toxicity prediction for the input molecule at the dose value based at least in part on the set of features characterizing the input molecule;
• identifying one or more molecular fragments of the input molecule that contribute to predicted toxicity across the sequence of dose values by processing a toxicity analysis tree generated based on the toxicity predictions for the input molecule to generate a toxicity score associated with each of the one or more molecular fragments, wherein processing the toxicity analysis tree comprises:
• for each molecular fragment of the one or more molecular fragments, identifying multiple nodes in the toxicity analysis tree that represent molecules which include the molecular fragment, generating a respective node-specific score for each node of the toxicity analysis tree based at least in part on a toxicity prediction for a molecule represented by the node and a size of the molecule represented by the node relative to a size of the input molecule, and aggregating the node-specific scores to generate the toxicity score for the molecular fragment;
• generating data defining a plurality of modified molecules, wherein each modified molecule is generated by replacing at least one of the identified molecular fragments with an alternative molecule fragment from a set of alternative molecule fragments;
• for each modified molecule in the plurality of modified molecules:
• for each dose value in the sequence of dose values, generating a toxicity prediction associated with the modified molecule at the dose value based at least in part on features characterizing the modified molecule;
• generating a set of toxicity dose-response curves for the input molecule and the set of modified molecules based on the toxicity predictions associated with the modified molecules;
• outputting data identifying the input molecule, the plurality of modified molecules, and the generated set of toxicity dose-response curves;
• and
• causing physical synthesis of a molecule selected based on the set of toxicity dose-response curves.

A non-transitory computer-readable storage medium storing executable instructions that when executed by a hardware processor, cause the hardware processor to perform steps comprising:

• obtaining data identifying:
• 1) an input molecule, and 2) a set of features characterizing the input molecule;
• for each dose value in a sequence of dose values, generating a toxicity prediction for the input molecule at the dose value based at least in part on the set of features characterizing the input molecule;
• identifying one or more molecular fragments of the input molecule that contribute to predicted toxicity across the sequence of dose values by processing a toxicity analysis tree generated based on the toxicity predictions for the input molecule to generate a toxicity score associated with each of the one or more molecular fragments, wherein processing the toxicity analysis tree comprises:
• for each molecular fragment of the one or more molecular fragments, identifying multiple nodes in the toxicity analysis tree that represent molecules which include the molecular fragment, generating a respective node-specific score for each node of the toxicity analysis tree based at least in part on a toxicity prediction for a molecule represented by the node and a size of the molecule represented by the node relative to a size of the input molecule, and aggregating the node-specific scores to generate the toxicity score for the molecular fragment;
• generating data defining a plurality of modified molecules, wherein each modified molecule is generated by replacing at least one of the identified molecular fragments with an alternative molecule fragment from a set of alternative molecule fragments;
• for each modified molecule in the plurality of modified molecules:
• for each dose value in the sequence of dose values, generating a toxicity prediction associated with the modified molecule at the dose value based at least in part on features characterizing the modified molecule;
• generating a set of toxicity dose-response curves for the input molecule and the set of modified molecules based on the toxicity predictions associated with the modified molecules;
• outputting data identifying the input molecule, the plurality of modified molecules, and the generated set of toxicity dose-response curve;
• and
• causing physical synthesis of a molecule selected based on the set of toxicity dose-response curves.