Optimizing liquid classes for a laboratory automation device

Patent metadata

Field	Value
Publication number	US-11249099-B2
Application number	US-201916359265-A
Country	US
Kind code	B2
Filing date	Mar 20, 2019
Priority date	Mar 22, 2018
Publication date	Feb 15, 2022
Grant date	Feb 15, 2022

Abstract

A liquid class including control parameters for controlling a laboratory automation device when handling a liquid of a liquid type assigned to the liquid class with a pipette tip of a pipette tip type assigned to the liquid class, and a method for optimizing the liquid class. The method can include initializing a genetic algorithm with a start liquid class by generating a set of liquid classes by modifying control parameters of the start liquid class. The method can further include optimizing the liquid class iteratively with the genetic algorithm by applying liquid classes from the set of liquid classes to the laboratory automation device, which performs at least one aspiration and dispensing measurement for each applied liquid class, as well as additional method acts.

First claim

The invention claimed is:

1. A method for optimizing a liquid class for a laboratory automation device, wherein the liquid class comprises control parameters for controlling the laboratory automation device when handling a liquid of a liquid type assigned to the liquid class with a pipette tip of a pipette tip type assigned to the liquid class, the method comprising:
initializing a genetic algorithm with a start liquid class by generating a set of liquid classes by modifying control parameters of the start liquid class;
optimizing the liquid class iteratively with the genetic algorithm by:
applying liquid classes from the set of liquid classes to the laboratory automation device, which laboratory automation device performs at least one aspiration and dispensing measurement for each applied liquid class, wherein during an aspiration and dispensing measurement, the laboratory automation device is controlled with the control parameters of the respective applied liquid class and a dispensed volume of a liquid of the liquid type is measured;
discarding liquid classes from the set of liquid classes responsive to a selection function that evaluates the at least one measured dispense volume for a liquid class;
adding liquid classes to the set of liquid classes, where the liquid classes that are added are generated by modifying liquid classes from the set of liquid classes; and
selecting an optimized liquid class from the set of liquid classes,

wherein the liquid class is additionally optimized with at least one of a simulation model, which comprises a geometric representation of at least a part of a dispensing and aspiration system of the laboratory automation device comprising a pipette tip of the pipette tip type, and a physical model, which comprises analytical equations modelling mechanical properties of at least a part of the dispensing and aspiration system,
wherein at least one of:
start control parameters for initializing the genetic algorithm are provided by at least one of the simulation model and the physical model, and
start control parameters for initializing at least one of the simulation model and the physical model are provided by the genetic algorithm,
wherein the liquid class is iteratively optimized with at least one of the simulation model and the physical model by:
simulating the part of the dispensing and aspiration system with the control parameters of the liquid class, wherein a liquid in the pipette tip is simulated based on physical properties of the liquid;
determining a dispensed volume from the at least one of the simulation model and the physical model; and

modifying the control parameters based on the dispensed volume; and
wherein the method further comprises controlling the laboratory automation device based on the control parameters.

2. The method of claim 1 , wherein:
the selection function selects liquid classes, where the at least one measured dispense volume differs not more than a threshold from a desired dispense volume; and/or
the aspiration and dispensing measurement is performed several times for an applied liquid class, several dispensed volumes are measured for the applied liquid class and the selection function selects liquid classes, where a variance of the several measured dispense volumes is smaller than a threshold.

3. The method of claim 1 , wherein:
modified control parameters for a modified liquid class are generated by modifying one control parameter of another liquid class; and/or
modified control parameters for a modified liquid class are generated by combining control parameters of two other liquid classes.

4. The method of claim 1 , wherein the control parameters comprise at least one of:
an aspiration speed, an aspiration acceleration and/or aspiration delay;
a dispensing speed, a dispensing acceleration and/or a dispensing delay;
a dipping speed and/or dipping acceleration of the pipette tip;
a pulling speed and/or pulling acceleration of the pipette tip;
an additional aspiration volume;
a dipping depth;
an airgap size; and
a subvolume size of an aspirated and/or dispensed volume.

5. The method of claim 1 , wherein:
the simulation model is based on finite elements and/or computational fluid dynamics; and/or
the part of the dispensing and aspiration system represented by the simulation model furthermore comprises a pump and a hose connecting the pump to the pipette tip; and/or
the geometric representation comprises physical properties of components of the dispensing and aspiration system.

6. The method of claim 1 , wherein the analytical equations model the liquid in the dispensing and aspiration system as an oscillating system having at least one damped mass connected to a spring.

7. A computer program stored on a non-transitory computer-readable medium for optimizing a liquid class for a laboratory automation device, which, when being executed by a processor, is configured to perform a method for optimizing a liquid class for a laboratory automation device, wherein the liquid class comprises control parameters for controlling the laboratory automation device when handling a liquid of a liquid type assigned to the liquid class with a pipette tip of a pipette tip type assigned to the liquid class, the method comprising:
initializing a genetic algorithm with a start liquid class by generating a set of liquid classes by modifying control parameters of the start liquid class; and
optimizing the liquid class iteratively with the genetic algorithm by:
applying liquid classes from the set of liquid classes to the laboratory automation device, which laboratory automation device performs at least one aspiration and dispensing measurement for each applied liquid class, wherein during an aspiration and dispensing measurement, the laboratory automation device is controlled with the control parameters of the respective applied liquid class and a dispensed volume of a liquid of the liquid type is measured;
discarding liquid classes from the set of liquid classes responsive to a selection function that evaluates the at least one measured dispense volume for a liquid class;
adding liquid classes to the set of liquid classes, where the liquid classes that are added are generated by modifying liquid classes from the set of liquid classes; and
selecting an optimized liquid class from the set of liquid classes,

wherein the liquid class is additionally optimized with at least one of a simulation model, which comprises a geometric representation of at least a part of a dispensing and aspiration system of the laboratory automation device comprising a pipette tip of the pipette tip type; and a physical model, which comprises analytical equations modelling mechanical properties of at least a part of the dispensing and aspiration system;
wherein at least one of:
start control parameters for initializing the genetic algorithm are provided by at least one of the simulation model and the physical model; and
start control parameters for initializing at least one of the simulation model and the physical model are provided by the genetic algorithm;
wherein the liquid class is iteratively optimized with at least one of the simulation model and the physical model by:
simulating the part of the dispensing and aspiration system with the control parameters of the liquid class, wherein a liquid in the pipette tip is simulated based on physical properties of the liquid;
determining a dispensed volume from the at least one of the simulation model and the physical model; and
modifying the control parameters based on the dispensed volume; and

controlling the laboratory automation device based on the control parameters.

8. An optimization system for optimizing a liquid class for a labor

Assignees

• Tecan Trading Ag

Inventors

• Wolter Anja
• Keller Michael

Classifications

• B01L3/0237

  Details of electronic control, e.g. relating to user interface · CPC title

• G01N35/00584

  Control arrangements for automatic analysers · CPC title

• G01N35/1016 Primary

  Control of the volume dispensed or introduced · CPC title

• B01L3/0293

  for liquids · CPC title