Enhancing flow cytometry discrimination with geometric transformation
In flow cytometry, particles ( 2 ) can be distinguished between populations ( 8 ) by combining n-dimensional parameter data, which may be derived from signal data from a particle, to mathematically achieve numerical results representative of an alteration ( 48 ). An alteration may include a rotational alteration, a scaled alteration, or perhaps even a translational alteration. Alterations may enhance separation of data points which may provide real-time classification ( 49 ) of signal data corresponding to individual particles into one of at least two populations.
1. A method of distinguishing particles having similar characteristics, comprising:
establishing a fluid stream;
entraining particles in the fluid stream, each individual particle belonging to one of at least two populations;
detecting a plurality of signals affiliated with individual particles in the fluid stream, the signals comprising signal data;
converting the signal data from the signals into n-dimensional parameter data, the n-dimensions including individual dimensions corresponding to respective signals in the plurality of signals, wherein the at least two populations of the individual particles overlap in one of the n-dimensions;
mathematically altering the n-dimensional parameter data by at least rotating the n-dimensional parameter data to increase spatial separation in the dimension having the overlap; and
real-time classifying the n-dimensional parameter data of each of the individual particles with the rotated n-dimensional parameter data; and
discriminating between the individual particles based upon the step of real-time classifying.
2. The method of claim 1 , further comprising: mathematically altering the n-dimensional parameter data by at least scaling the n-dimensional parameter data; and real-time classifying the n-dimensional parameter data of each of the individual particles into one of the at least two populations based upon at least the scaled and rotated altered parameter data.
3. The method of claim 2 , further comprising: simultaneously processing the n-dimensional parameter data to alter the data by both scaling and rotation.
4. The method of claim 1 , further comprising: mathematically altering the n-dimensional parameter data by at least translating the n-dimensional parameter data; and real-time classifying the n-dimensional parameter data of each of the individual particles into one of the at least two populations based upon at least the translated and rotated n-dimensional parameter data.
5. The method of claim 1 , further comprising: simultaneously processing the n-dimensional parameter data to mathematically achieve numerical results of the rotational alteration in addition to a scaled alteration and a translation alteration.
6. The method of claim 1 further comprising: generating a graphical representation of the at least two populations of the rotated n-dimensional parameter data.
7. The method of claim 6 , wherein the step of generating a graphical representation of the at least two populations of the rotated n-dimensional parameter data comprises plotting the rotated n-dimensional parameter data in a Cartesian coordinate system.
8. The method of claim 7 , wherein plotting the rotated n-parameter data in a Cartesian coordinate system further comprises the step of plotting of the n-dimensional parameter data having the overlap on an axis.
9. The method of claim 1 , further comprising: visually distinguishing the at least two populations of the n-dimensional parameter data by: graphically providing elliptical shaped populations of each of the at least two populations having non-orthogonal angles of inclination to at least one axis; and rotating the elliptical shaped populations orthogonal to at least one axis.
10. A method of claim 1 , wherein the particles comprise sperm and wherein at least one of the two populations comprises an X-bearing sperm population or a Y-bearing sperm population.
11. The method of claim 1 , wherein discriminating further comprises sorting individual particles based on their real-time classification.
12. The method of claim 1 , wherein the n-dimensional parameter data comprises 2-dimensional parameter data.
13. The method of claim 12 , wherein mathematically altering the 2-dimensional parameter data by at least rotating the 2-dimensional parameter data to increase the spatial separation in the dimension having the overlap further comprises rotating the 2-dimensional parameter data such that a major axis of each of the at least two populations of the individual particles is rotated to an orthogonal orientation with one axis of a Cartesian coordinate system.