Self-tuning, biologically modeled sorter
AssigneeXy, LLC
InventorsDaniel N. Fox, George C. Malachowski, Matthias J. G. Ottenberg, Tidhar Sadeh
Life of this patent
Estimating the term…
Abstract
Disclosed is a self-tuning flow cytometer that uses a mathematical model to perform sort decisions that is based upon the biological response of the particular types of cells that are being sorted. In one embodiment, statistical calculations of the likelihood of an event belonging to a certain population are used to make the sort decisions. Automated self-tuning processes are used to optimize the operating parameters of the flow cytometer to achieve a selected purity with higher yield at optimal sorting speeds. The fully automated processes minimize user input and allows the user to select a desired purity while maximizing yield.
Claims9
A method of generating a sort decision signal for sorting a cell of a plurality of cells in a sort logic controller of a flow cytometer comprising:
- calculating statistical event probability data in said sort logic controller of said flow cytometer comprising:
- detecting a biological response of said cell in response to illumination by a source of electromagnetic radiation using optical detectors;
- calculating a statistical probability that said cell belongs to a predetermined population of one of at least two overlapping Gaussian distribution functions based upon analysis of said biological response using an event probability calculator in said sort logic controller comprising;
- deconvolving said two Gaussian distribution functions;
- calculating mean values and standard deviation values for said two Gaussian distribution functions;
- using said mean values and said standard deviation values in a probability distribution function to derive said statistical event probability data for said at least two overlapping Gaussian distribution functions;
- comparing values of said statistical event probability data to determine a most likely population of said at least two overlapping Gaussian distribution functions to which said cell belongs;
- calculating droplet probability data in said sort logic controller comprising:
- determining a location of said cell in a droplet that is formed in said flow cytometer;
- comparing said location of said cell in said droplet with probability data indicating a probability that said cell is present in said predetermined droplet using a droplet probability calculator;
- calculating droplet desirability data for said droplet comprising:
- combining said statistical event probability data and said droplet probability data using a droplet desirability calculator in said sort logic controller of said flow cytometer;
- generating a sort decision signal by comparing said droplet desirability data with an internally set threshold using a sort decision calculator in said sort logic controller of said flow cytometer.
Dependencies are inferred from the claim text.