Statistical outlier detection for gene expression microarray data
AssigneeSAS Institute Inc.
InventorsTzu-Ming Chu, Russell D. Wolfinger
Life of this patent
Estimating the term…
Abstract
In accordance with the disclosure below, a computer-implemented method and system are provided for detecting outliers in microarray data. A mixed linear statistical model is used to generate predictions based upon the received microarray data. Residuals are generated by subtracting model-based predictions from the original microarray sample data. Statistical tests are performed for residuals by adding covariates to the mixed model and testing their significance. Data from the microarrays are designated as outliers based upon the tested significance.
Claims27
A computer-implemented method for detecting outliers in microarray data, said method comprising the steps of:
- receiving the microarray data, said microarray data containing data values indicative of at least one characteristic associated with processed gene samples;
- using a mixed linear statistical model to generate predictions based upon the received microarray data;
- generating residuals based upon the predictions and the received microarray data;
- performing a statistical test for at least one generated residual by adding covariates to the mixed linear mathematical model and testing significance of the covariates;
- and
- designating a data value within the received microarray data as an outlier based upon the tested significance.
A computer-implemented method for detecting outliers in microarray data, said method comprising the steps of:
- receiving the microarray data, said microarray data containing data values indicative of at least one characteristic associated with processed gene samples;
- using a statistical model to generate predictions based upon the received microarray data, wherein the statistical model includes multiple degrees of freedom;
- generating residuals by comparing the predictions with the received microarray data;
- performing a statistical test for at least one generated residual by adding covariates to the mathematical model and testing significance of the covariates;
- and
- designating a data value within the received microarray data as an outlier based upon the tested significance.
A computer-implemented method for detecting outliers in microarray data, said method comprising the steps of:
- receiving the microarray data, said microarray data containing data values indicative of at least one characteristic associated with processed gene samples;
- using a mixed mathematical model to generate predictions based upon the received microarray data;
- generating residuals by comparing the predictions with the received microarray data;
- and
- designating a data value within the received microarray data as an outlier based upon the magnitude of the generated residuals.
A computer-implemented apparatus for detecting outliers in microarray data, comprising:
- means for receiving the microarray data, said microarray data containing data values indicative of at least one characteristic associated with processed gene samples;
- means for using a mixed mathematical model to generate predictions based upon the received microarray data;
- means for generating residuals by comparing the predictions with the received microarray data;
- and
- means for designating a data value within the received microarray data as an outlier based upon the magnitude of the generated residuals.
A computer-implemented statistical system for analyzing microarray data, said microarray data containing data values indicative of at least one characteristic associated with processed gene samples, said system comprising:
- a mixed linear statistical model that generates predictions based upon the received microarray data,
- wherein residuals are generated based upon the predictions and the received microarray data;
- a statistical test to be performed for at least one generated residual by adding covariates to the mixed linear mathematical model,
- wherein the significance of the covariates are tested, wherein a data value within the received microarray data is designated as an outlier based upon the tested significance.
Dependencies are inferred from the claim text.