Visualization, comparative analysis, and automated difference detection for large multi-parameter data sets

1. A computer-implemented method of visualizing differences between n-dimensional data sets, the computer-implemented method comprising:

under control of one or more processing devices,

receiving at least one threshold for frequency difference gating received data sets;

receiving a first data set of n-dimensional data including a first plurality of flow cytometry events, corresponding to a first sample, in a plurality of first dimensions;

receiving a second data set of n-dimensional data including a second plurality of flow cytometry events, corresponding to a second sample for cross-sample comparison with the first sample, in at least the plurality of first dimensions;

concatenating the first data set and the second data set to obtain a concatenated data set;

performing dimensionality reduction on the concatenated data set to obtain a mapping of events of the concatenated data set in a multivariate space defined by second dimensions different from and fewer than the plurality of first dimensions;

identifying a frequency difference gate defining a population of events based at least in part on frequency difference gating performed on the mapping of events, the gate identifying a region in the multivariate space where a frequency of events from the first data set is different than a frequency of events from the second data set according to the at least one threshold; and

generating a visualization from the frequency difference gated data for display via a display device, the visualization including a representation of events from the first data set and the second data set included in the population defined by the frequency difference gate, the visualization showing regions in the multivariate space where the frequency of events from the first data set is different than the frequency of events from the second data set according to the at least one threshold.

2. The computer-implemented method of claim 1 , wherein the first and second data sets comprise multi-parameter cell sample data.

3. The computer-implemented method of claim 1 , wherein the at least one threshold includes an upper threshold identifying one or more regions classified as having a greater frequency of events from the first data set than the second data set.

4. The computer-implemented method of claim 1 , wherein the at least one threshold includes a lower threshold identifying one or more regions having a greater frequency of events from the second data set than the first data set.

5. The computer-implemented method of claim 1 , wherein the at least one threshold includes mid-range boundaries identifying one or more regions having a similar frequency of events as between the first data set and the second data set.

6. The computer-implemented method of claim 1 , wherein performing frequency difference gating comprises:

generating a multi-dimensional histogram with a plurality of bins per dimension according to a bivariate frequency estimate of each of a plurality of defined distributions within the first data set and the second data set.

7. A system comprising:

one or more processing devices; and

a computer-readable storage medium comprising instructions that are configured to, when executed by the one or more processing devices, cause the system to:

receive at least one threshold for frequency difference gating received data sets;

receive a first data set of n-dimensional data including a first plurality of flow cytometry events, corresponding to a first sample, in a plurality of first dimensions;

receive a second data set of n-dimensional data including a second plurality of flow cytometry events, corresponding to a second sample for cross-sample comparison with the first sample, in at least the plurality of first dimensions;

concatenate the first data set and the second data set to obtain a concatenated data set;

perform dimensionality reduction on the concatenated data set to obtain a mapping of events of the concatenated data set in a multivariate space defined by second dimensions different from and fewer than the plurality of first dimensions;

identify a frequency difference gate defining a population of events based at least in part on frequency difference gating performed on the mapping of events, the gate identifying a region in the multivariate space where a frequency of events from the first data set is different than a frequency of events from the second data set according to the at least one threshold; and

display a visualization including a representation of events from the first data set and the second data set included in the population defined by the frequency difference gate, the visualization showing regions in the multivariate space where the frequency of events from the first data set is different than the frequency of events from the second data set according to the at least one threshold.

8. The system of claim 7 , wherein the instructions are further configured to cause the system to:

adjust the at least one threshold in response to user input; and

adjust the visualization based on the adjusted at least one threshold.

9. The system of claim 7 , wherein the at least one threshold comprises a plurality of thresholds.

10. The system of claim 7 , wherein displaying the visualization comprises displaying the regions with color coding indicating relative differences in event frequency between the first and second data sets, based at least in part on the frequency difference gate.

11. The system of claim 7 , wherein the instructions are further configured to cause the system to:

generate a third data set based on at least one gate defined by a user through the visualization.

12. The system of claim 7 , wherein the first sample comprises a control sample.

13. The system of claim 12 , wherein the control sample corresponds to healthy matter.

14. The system of claim 12 , wherein the control sample corresponds to cancerous matter.

15. The system of claim 7 , wherein the at least one threshold includes at least one of:

an upper threshold identifying one or more regions classified as having a greater frequency of events from the first data set than the second data set;

a lower threshold identifying one or more regions having a greater frequency of events from the second data set than the first data set; or

two thresholds corresponding to mid-range boundaries identifying one or more regions having a similar frequency of events as between the first data set and the second data set.

16. The system of claim 7 , wherein the computer-readable storage medium comprises instructions that are configured to, when executed by the one or more processing device, cause the system to identify the frequency difference gate by at least:

generating a multi-dimensional histogram with a plurality of bins per first dimension according to a bivariate frequency estimate of each of a plurality of defined distributions within the first data set and the second data set.

17. The system of claim 16 , wherein identifying the frequency difference gate further comprises:

normalizing the multi-dimensional histogram by event count.

18. The system of claim 17 , wherein identifying the frequency difference gate further comprises:

generating difference histograms for each element in the normalized histogram.

19. The system of claim 18 , wherein identifying the frequency difference gate further comprises:

bi-normalizing the difference histograms.

20. The system of claim 19 wherein displaying the visualization comprises:

rendering a heat map of the bi-normalized difference histograms.