Method and apparatus for multi-parameter data analysis

1. A method for automated high-content screening (HCS) of a plurality of samples comprising cells and subjected to sample treatment conditions in an assay, wherein the plurality of samples comprise both samples subjected to a same treatment condition and samples subjected to different sample treatment conditions, the method comprising:

processing data from each of two or more images of the plurality of samples to obtain a multi-parameter data set for each image;

for samples subjected to the same sample treatment condition, obtaining inter-sample measurements based on distributions of different parameters in the same sample under the same sample treatment condition;

for samples subjected to different sample treatment conditions, obtaining intra-sample measurements based on distributions of same parameter under the different sample treatment conditions;

determining, based on both inter-sample and intra-sample measurements, correlated parameters in the respective multi-parameter data sets, wherein parameters from the multi-parameter data sets are graphically visualized in a phenotype map generated based on the inter-sample and intra-sample measurements and compared to determine the correlated parameters, and wherein correlation is to be determined using a Kolmogorov-Smirnov (KS) distance measurement analysis of parameters from the multi-parameter data sets;

creating an analysis parameter set by removing at least one of the correlated parameters from the multi-parameter data sets to form the analysis parameter set;

applying the analysis parameter set to identify one or more cellular phenotypes of the samples from the multi-parameter data sets; and

providing an identification, in the plurality of samples, of one or more cells that undergo changes in cellular phenotypes.

2. The method of claim 1 , further comprising identifying one or more phenotypes from a plurality of multi-parameter data sets and comparing respective phenotypes from the multi-parameter data sets to identify variations therebetween.

3. The method of claim 1 , further comprising forming one or more respective analysis parameter sets for a plurality of multi-parameter data sets, and comparing the analysis parameter sets to determine whether or not a correlation relationship is maintained between the multi-parameter data sets.

4. The method of claim 1 , wherein correlation is determined by the KS distance measurement analysis made between the phenotypes in a predetermined multi-parameter data analysis set.

5. The method of claim 4 , wherein the KS distance measurement analysis includes a comparison of the frequency distribution of data within two data sets in the predetermined multi-parameter data analysis set.

6. The method of claim 1 , wherein modifying correlated parameter values within the predetermined multi-parameter data analysis set comprises removing one of a pair of parameters whose correlation exceeds a predetermined threshold value from the analysis parameter set.

7. A tangible non-transient computer program storage product comprising machine instructions operable to configure a data processing apparatus to:

process data from each of two or more images of a plurality of samples to obtain a multi-parameter data set for each image;

for samples subjected to a same sample treatment condition, obtain inter-sample measurements based on distributions of different parameters in the same sample under the same sample treatment condition;

for samples subjected to different sample treatment conditions, obtain intra-sample measurements based on distributions of same parameter under the different sample treatment conditions;

determine, based on both inter-sample and intra-sample measurements, correlated parameters in the respective multi-parameter data sets, wherein parameters from the multi-parameter data sets are graphically visualized in a phenotype map generated based on the inter-sample and intra-sample measurements and compared to determine the correlated parameters, and wherein correlation is to be determined using a Kolmogorov-Smirnov (KS) distance measurement analysis of parameters from the multi-parameter data sets;

create an analysis parameter set by removing at least one of the correlated parameters from the multi-parameter data sets to form the analysis parameter set;

apply the analysis parameter set to identify one or more cellular phenotypes of the samples from the multi-parameter data sets; and

provide an identification, in the plurality of samples, of one or more cells that undergo changes in cellular phenotypes.

8. A system for automated high-content screening (HCS) of a plurality of samples comprising cells and subjected to sample treatment conditions in an assay, wherein the plurality of samples comprise both samples subjected to a same treatment condition and samples subjected to different sample treatment conditions, the system comprising:

an imager for obtaining two or more images of the plurality of samples, each image represented by a respective multi-parameter data set; and

an image processor that is operable to:

process data from each of the two or more images to obtain respective ones of the multi-parameter data sets;

for samples subjected to the same sample treatment condition, obtain inter-sample measurements based on distributions of different parameters in the same sample under the same sample treatment condition;

for samples subjected to different sample treatment conditions, obtain intra-sample measurements based on distributions of same parameter under the different sample treatment conditions;

determine, based on both inter-sample and intra-sample measurements, correlated parameters in the respective multi-parameter data sets, wherein parameters from the multi-parameter data sets are graphically visualized in a phenotype map generated based on the inter-sample and intra-sample measurements and compared to determine the correlated parameters, and wherein correlation is to be determined using a Kolmogorov-Smirnov (KS) distance measurement analysis of parameters from the multi-parameter data sets;

create an analysis parameter set by removing at least one of the correlated parameters from the multi-parameter data sets to form the analysis parameter set;

apply the analysis parameter set to identify one or more cellular phenotypes of the samples from the multi-parameter data sets; and

provide an identification, in the plurality of samples, of one or more cells that undergo changes in cellular phenotypes.

9. The system of claim 8 , wherein the image processor is further operable to identify one or more phenotypes from a plurality of multi-parameter data sets and compare respective phenotypes from the multi-parameter data sets to identify variations therebetween.

10. The system of claim 8 , wherein the image processor is further operable to form two or more respective analysis parameter sets for a plurality of multi-parameter data sets, and compare the analysis parameter sets to determine whether or not a correlation relationship is maintained between the multi-parameter data sets.

11. The system of claim 8 , wherein the image processor is further operable to determine correlation by using the KS distance measurement analysis made between the parameters in a predetermined multi-parameter data analysis set.

12. The system of claim 11 , wherein the KS distance measurement analysis includes a comparison of the frequency distribution of data within two data sets in the predetermined multi-parameter data analysis set.

13. The system of claim 8 , wherein the image processor is further operable to modify correlated parameter values within a predetermined multi-parameter data analysis set by removing one of a pair of parameters whose correlation exceeds a predetermined threshold value from the analysis parameter set.

14. The system of claim 8 , wherein the two or more images comprise cellular images and the imager comprises a cellular imager.

15. The system of claim 14 , wherein each multi-parameter data set includes one or more parameters of: nuclear area; length; diameter; perimeter; ellipticity; and weighted moment of inertia (WMOI), form, and distribution.

16. The system of claim 14 , wherein the intra-sample measurements are used in weighting and prioritizing data produced from the inter-sample measurements to produce a single phenotype score measure summarizing a cumulative difference between two cellular populations across a range of measured parameters.

17. The system of claim 8 , wherein the image processor is further configured to produce one or more phenotype maps graphically representing phenotypic signatures based on intra-sample measurements and inter-sample measurements.

18. The system of claim 17 , wherein the inter-sample measurements and the intra-sample measurements comprise population comparator measures.

19. The system of claim 18 , wherein the population comparator measures comprise Kolmogorov-Smirnov (KS) distance measurements.

20. The system of claim 8 , wherein the intra-sample measurements are utilized to measure changes in the inter-sample measurements to provide phenotype characterization data.

21. The system of claim 8 , wherein the sample is subjected to a drug treatment.

22. The system of claim 8 , wherein the sample comprises a cellular population and the two or more images comprises a microscope image of the cellular population.

23. The system of claim 8 , wherein identifying cells that undergo changes in cellular phenotypes comprises detecting changes in interrelationships of correlated parameters based on inter sample measurement.

24. The system of claim 8 , the image processor is further operable to generate, based on a combination of inter-sample and intra-sample measurements, phenotype maps graphically representing phenotypic signatures.