Models for combinatorial perturbations of living biological systems

A method for outputting values for a plurality of state variables, wherein each of said method steps is performed on a processor, comprising:

• receiving a set of initial state values for a plurality of state variables y that indicate relevant properties of a living biological system;
• receiving model data that includes values for a plurality of automatically determined, numerical, constant interaction factors wij between pairs of state variables;
• receiving a set of trial values for a plurality of perturbation variables u that indicate factors that might affect one or more of the relevant properties of the biological system;
• and
• determining a change over time in a value for a particular state variable yi, of the plurality of state variables based, at least in part, on a non-linear transformation of a sum of the trial value for the perturbation variable ui that affects yi added to a sum of all non-zero values for a product of wij and yj over all state variables in the plurality of state variables.

A non-transitory computer-readable medium carrying one or more sequences of instructions for modeling a living biological system, wherein execution of the one or more sequences of instructions by one or more processors causes an apparatus to perform the steps of:

• receiving state data that indicates a plurality of state variables y that indicate relevant properties of a living biological system;
• receiving perturbation data that indicates a plurality of perturbation variables u that indicate factors that might affect one or more of the relevant properties of the biological system;
• receiving measurable data that indicates a measurable subset Ω of the plurality of state variables y, wherein a measureable variable of Ω is a state variable for which a value can be derived based on a measurement;
• receiving measurement-based values Y for the measurable variables in Ω;
• determining model constants automatically based on the measured values Y, including determining a numerical value for a constant interaction factor wij between a first state variable yi and a second state variable yj such that a change in a value of yi over time depends, at least in part, on a non-linear transformation of a sum of a value of a perturbation variable ui that affects yi added to a sum of all non-zero values for a product of wij and yj for all state variables in the plurality of state variables;
• determining model data that indicates the model constants, the plurality of state variables y and the plurality of perturbation variables u for a model that predicts a temporal change in yi based on values for the plurality of state variables y and for a perturbation variable ui, and
• writing the model data to a computer-readable medium.

A non-transitory computer-readable medium carrying one or more sequences of instructions for modeling a living biological system, wherein execution of the one or more sequences of instructions by one or more processors causes the one or more processors to perform the steps of:

• receiving a set of initial state values for a plurality of state variables y that indicate relevant properties of a living biological system;
• receiving model data that includes values for a plurality of automatically determined, numerical, constant interaction factors wij between pairs of state variables;
• receiving a set of trial values for a plurality of perturbation variables u that indicate factors that might affect one or more of the relevant properties of the biological system;
• and
• determining a change over time in a value for a particular state variable yi of the plurality of state variables based, at least in part, on a non-linear transformation of a sum of the trial value for the perturbation variable ui that affects yi added to a sum of all non-zero values for a product of wij and yj over all state variables in the plurality of state variables.

A method for screening drugs for treating a biological system, wherein each of said method steps is performed on a processor, comprising:

• receiving a set of initial state values for a plurality of state variables y that indicate relevant properties of a living biological system;
• receiving model data that includes values for a plurality of automatically determined, numerical, constant interaction factors wij between pairs of state variables;
• receiving a set of trial values for a set of one or more drugs that might affect one or more of the relevant properties of the biological system;
• determining a change over time in a value for a particular state variable yi of the plurality of state variables based, at least in part, on a non-linear transformation of a sum of the set of trial values that affect yi added to a sum of all non-zero values for a product of wij and yj over all state variables in the plurality of state variables;
• and
• determining efficacy of the set of trial values for the set of one or more drugs in treating the biological system based on the change over time in the particular state variable.

A system for modeling living biological systems, comprising:

• one or more computer-readable media configured to store data indicating:
• a plurality of state variables y that indicate relevant properties of a living biological system;
• a plurality of perturbation variables u that indicate factors that might affect one or more of the relevant properties of the biological system;
• a measurable subset Ω of the plurality of state variables y, wherein a measureable variable of Ω is a state variable for which a value can be derived based on a measurement;
• and
• a set of one or more processors configured to perform the following:
• receive measurement-based values Y for the measurable variables in Ω;
• determine model constants automatically on a processor based on the measured values Y, including determining a numerical value for a constant interaction factor wij between a first state variable yi and a second state variable yj such that a change in a value of yi over time depends, at least in part, on a non-linear transformation of a sum of a value of a perturbation variable ui that affects yi added to a sum of all non-zero values for a product of wij and yj for all state variables in the plurality of state variables;
• determine model data that indicates the model constants, the plurality of state variables y and the plurality of perturbation variables u for a model that predicts a temporal change in yi based on values for the plurality of state variables y and for a perturbation variable u1 and
• write model data to a computer-readable medium.

A system for outputting values for a plurality of state variables comprising a set of one or more processors configured to perform the following:

• receiving a set of initial state values for a plurality of state variables y that indicate relevant properties of a living biological system;
• receiving model data that includes values for a plurality of automatically determined, numerical, constant interaction factors wij between pairs of state variables;
• receiving a set of trial values for a plurality of perturbation variables u that indicate factors that might affect one or more of the relevant properties of the biological system;
• and
• determining a change over time in a value for a particular state variable yi of the plurality of state variables based, at least in part, on a non-linear transformation of a sum of the trial value for the perturbation variable ui that affects yi added to a sum of all non-zero values for a product of wij and yj over all state variables in the plurality of state variables.

A system for screening drugs for treating a biological system comprising a set of one or more processors configured to perform the following:

• receiving a set of initial state values for a plurality of state variables y that indicate relevant properties of a living biological system;
• receiving model data that includes values for a plurality of automatically determined, numerical, constant interaction factors wij between pairs of state variables;
• receiving a set of trial values for a set of one or more drugs that might affect one or more of the relevant properties of the biological system;
• determining a change over time in a value for a particular state variable yi of the plurality of state variables based, at least in part, on a non-linear transformation of a sum of the set of trial values that affect yi added to a sum of all non-zero values for a product of wij and yj over all state variables in the plurality of state variables;
• and
• determining efficacy of the set of trial values for the set of one or more drugs in treating the biological system based on the change over time in the particular state variable.