Systems, methods, and media for molecule design using machine learning mechanisms

A method for proposing molecules having desired functional property values, comprising:

• forming a first training set for a neural network by:
• selecting features of building block molecules to produce selected features and functional properties of a first plurality of proposed molecules that comprise the building block molecules to produce selected functional properties;
• accessing a library that identifies known molecules Aknown made from the building block molecules;
• forming a representation of the library in which each known molecule Aknown is shown as a matrix Bknown of representations of the building block molecules;
• forming a matrix T of the values V of the selected features of all of the building block molecules in the library or all of features of the building block molecules;
• multiplying each matrix Bknown by the matrix T to form a matrix Cknown of features;
• linearizing each matrix Cknown to form vectors Dknown;
• and
• determining, for each vector Dknown, known values Pknown for corresponding known molecules Aknown that comprise the selected functional properties;
• training the neural network using at least the vectors Dknown of the first training set using a hardware processor to predict values Pknown of the selected function properties to produce a trained neural network;
• proposing a first plurality of proposed molecules Aproposed, and predict values Pproposed of the selected functional properties of the proposed molecules Aproposed using the trained neural network;
• synthesizing selected molecules of the first plurality of proposed molecules Aproposed to form a first plurality of synthesized molecules Asynth using a computerized synthesizer;
• receiving first measured functional property values of the first plurality of synthesized molecules Asynth;
• and
• adding data regarding selected molecules of the first plurality of synthesized molecules Asynth to the library used to form the first training set to form a second training set and retrain the neural network using the second training set.

A non-transitory computer-readable medium containing computer executable instructions that, when executed by a processor, cause the processor to perform a method proposing molecules having desired functional property values, the method comprising:

• forming a first training set for a neural network by:
• selecting features of building block molecules to produce selected features and functional properties of a first plurality of proposed molecules that comprise the building block molecules to produce selected functional properties;
• accessing a library that identifies known molecules Aknown made from the building block molecules;
• forming a representation of the library in which each known molecule Aknown is shown as a matrix Bknown of representations of the building block molecules;
• forming a matrix T of the values V of the selected features of all of the building block molecules in the library or all of features of the building block molecules;
• multiplying each matrix Bknown by the matrix T to form a matrix Cknown of features;
• linearizing each matrix Cknown to form vectors Dknown;
• and
• determining, for each vector Dknown, known values Pknown for corresponding known molecules Aknown that comprise the selected functional properties;
• training the neural network using at least the vectors Dknown of the first training set using a hardware processor to predict values Pknown of the selected function properties to produce a trained neural network;
• proposing a first plurality of proposed molecules Aproposed, and predict values Pproposed of the selected functional properties of the proposed molecules Aproposed using the trained neural network;
• synthesizing selected molecules of the first plurality of proposed molecules Aproposed to form a first plurality of synthesized molecules Asynth using a computerized synthesizer;
• receiving first measured functional property values of the first plurality of synthesized molecules Asynth;
• and
• adding data regarding selected molecules of the first plurality of synthesized molecules Asynth to the library used to form the first training set to form a second training set and retrain the neural network using the second training set.