Methods for the analysis of dissociation melt curve data
Methods are provided that operate on raw dissociation data and dissociation curves to generate calibrations of the detected data and to further improve analysis of the data. The data can be taken from each support region of a multi-region platform, for example, from each well of a multi-well plate. Each support region can be loaded with portions of the same sample. In some embodiments, a dissociation curve correction can be calibrated for the sample, prior to a run of an experiment using such sample. In some embodiments, a method is provided for generating a melting transition region of dissociation curves that show the melting characteristics of the sample. In some embodiments, dye temperature dependence correction can be performed on the dissociation curve data to further improve analysis. In some embodiments, a feature vector can be derived from the melt data, and the feature vector can be used to further improve genotyping analysis of the dissociation curves.
1. A method of a biological analysis system for analyzing melt curve data, wherein the biological analysis system includes a processor and a display, the method comprising:
generating, by the processor, a calibration set of melt curve data for a calibration nucleic acid sample, wherein the calibration set is generated from subjecting the calibration nucleic acid sample deposited in a first plurality of support regions to conditions sufficient to denature nucleic acid replicates in the calibration nucleic acid sample;
generating, by the processor, an experimental set of melt curve data for at least one test nucleic acid sample, wherein the experimental set is generated from subjecting the at least one test nucleic acid sample deposited in a second plurality of support regions to conditions sufficient to denature nucleic acids in the at least one test nucleic acid sample;
correcting, by the processor, assay system noise in the experimental set of melt curve data using the calibration set of melt curve data, wherein the assay system noise is caused by at least thermal non-uniformity between support regions from the first plurality of support regions and the second plurality of support regions, and wherein correcting the experimental set of melt curve data for assay system noise using the calibration set of melt curve data further comprises:
calculating, for the calibration set of melt curve data, a correction value comprising a difference between data from at least two support regions among the first plurality of support regions in which the calibration nucleic acid sample is deposited, and
applying, for the experimental set of melt curve data, the calculated correction value to data from at least one of the second plurality of support regions in which the at least one test nucleic acid sample is deposited;
scaling the corrected experimental set of melt curve data over an estimated temperature range;
fitting the scaled corrected experimental set of melt curve data to an estimated asymptote for a low temperature region of a melting region of the melt curve data;
clustering the experimental set of melt curve data based on the steps of correcting, scaling and fitting; and
displaying, on the display, the corrected experimental set of melt curve data to a user.
2. The method of claim 1 , wherein the method further comprises creating difference data from the experimental melt curve data for a plurality of nucleic acid samples, wherein the experimental melt curve data for one of the plurality of nucleic acid samples is selected as a reference, and the experimental melt curve data for the remaining nucleic acid samples are subtracted from the reference nucleic acid sample to create the difference melt curve data.
3. The method of claim 2 , wherein feature vectors for one or more of the plurality of samples are generated from the difference melt curve data.
4. The method of claim 2 , further comprising genotyping the plurality of nucleic acid samples based on the generated difference melt curve data.
5. The method of claim 3 , wherein clustering the experimental set of melt curve data further comprises:
clustering the experimental melt curve data based on the difference melt curve data.
6. The method of claim 3 , wherein the feature vector for at least one sample from among the plurality of nucleic acid samples comprises one or more of a delta max, a melt temperature at delta max, and a sum of an absolute difference, wherein the delta max comprises a difference between a peak value for the at least one nucleic acid sample and the reference nucleic acid sample, the melt temperature at delta max comprises the melt temperature at the peak value, and the sum of the absolute difference comprises the area under the sample peak.
7. The method of claim 1 , wherein the correcting is based on a derivative form of the experimental melt curve data.
8. The method of claim 1 , wherein the calibration set of melt curve data is generated based on detected changes to an indicator exhibited by the plurality of support regions in which the calibration nucleic acid sample is deposited, the detected changes occurring during the subjecting to conditions sufficient to denature the nucleic acid replicates in the calibration nucleic acid sample.
9. A biological analysis system for analyzing melt curve data, the biological analysis system comprising:
a processor configured to:
generate a calibration set of melt curve data for a calibration nucleic acid sample, wherein the calibration set is generated from subjecting the calibration nucleic acid sample deposited in a first plurality of support regions to conditions sufficient to denature nucleic acid replicates in the calibration nucleic acid sample;
generate an experimental set of melt curve data for at least one test nucleic acid sample, wherein the experimental set is generated from subjecting the at least one test nucleic acid sample deposited in a second plurality of support regions to conditions sufficient to denature nucleic acids in the at least one test nucleic acid sample;
correct assay system noise in the experimental set of melt curve data using the calibration set of melt curve data, wherein the assay system noise is caused by at least thermal non-uniformity between support regions from the first plurality of support regions and the second plurality of support regions, and wherein correcting the experimental set of melt curve data for assay system noise using the calibration set of melt curve data further comprises:
calculating, for the calibration set of melt curve data, a correction value comprising a difference between data from at least two support regions among the first plurality of support regions in which the calibration nucleic acid sample is deposited, and
applying, for the experimental set of melt curve data, the calculated correction value to data from at least one of the second plurality of support regions in which the at least one test nucleic acid sample is deposited;
scale the corrected experimental set of melt curve data over an estimated temperature range;
fit the scaled corrected experimental set of melt curve data to an estimated asymptote for a low temperature region of a melting region of the melt curve data;
cluster the experimental set of melt curve data based on the steps of correcting, scaling and fitting; and
a display configured to:
display the corrected experimental set of melt curve data to a user.
10. The biological analysis system of claim 9 , wherein the processor is further configured to create difference melt curve data from the experimental melt curve data for a plurality of nucleic acid samples, wherein the experimental melt curve data for one of the plurality of nucleic acid samples is selected as a reference, and the experimental melt curve data for the remaining nucleic acid samples are subtracted from the reference nucleic acid sample to create the difference melt curve data.
11. The biological analysis system of claim 10 , wherein feature vectors are generated from the difference melt curve data.
12. The system of claim 10 , wherein the processor is configured to genotype the plurality of nucleic acid samples based on the generated difference melt curve data.
13. The biological analysis system of claim 11 , wherein clustering the experimental set of melt curve data comprises:
clustering the experimental melt curve data based on the difference melt curve data.
14. The system of claim 11 , wherein the feature vector for at least one sample from among the plurality of nucleic acid samples comprises one or more of a delta max, a melt temperature at delta max, and a sum of an absolute difference, wherein the delta max comprises a difference between a peak value for the at least one nucleic acid sample and the reference nucleic acid sample, the melt temperature at delta max comprises the melt temperature at the peak value, and the sum of the absolute difference comprises the area under the sample peak.
15. The biological analysis system of claim 9 , wherein the processor is configured to correct the experimental set of melt curve data for assay system noise using a derivative form of the experimental melt curve data.
16. The system of claim 9 , wherein the calibration set of melt curve data is generated based on detected changes to an indicator exhibited by the plurality of support regions in which the calibration nucleic acid sample is deposited, the detected changes occurring during the subjecting to conditions sufficient to denature the nucleic acid replicates in the calibration nucleic acid sample.
17. A method of a biological analysis system for analyzing melt curve data, wherein the biological analysis system includes a processor and a display, the method comprising:
generating a calibration set of melt curve data for a calibration nucleic acid sample, wherein the calibration set of melt curve data comprises fluorescence measurements collected by subjecting the calibration nucleic acid sample deposited in a first plurality of support regions to conditions sufficient to denature nucleic acid replicates in the calibration nucleic acid sample;
generating an experimental set of melt curve data for at least one test nucleic acid sample, wherein the experimental set of melt curve data comprises fluorescence measurements collected by subjecting the at least one test nucleic acid sample deposited in a second plurality of support regions to conditions sufficient to denature nucleic acids in the at least one test nucleic acid sample;
correcting assay system noise in the experimental set of melt curve data using a correction value calculated from the calibration set of melt curve data, wherein the assay system noise is caused by at least thermal non-uniformity between support regions from the first plurality of support regions and the second plurality of support regions;
scaling the corrected experimental set of melt curve data over an estimated temperature range;
fitting the scaled experimental set of melt curve data to an estimated asymptote for a low temperature region of a melting region of the melt curve data;
clustering the experimental set of melt curve data based on the steps of correcting, scaling and fitting; and
displaying, on the display, the corrected experimental set of melt curve data to a user.
18. The method of claim 17 , wherein correcting the experimental set of melt curve data comprises:
calculating, for the calibration set of melt curve data, the correction value, the correction value comprising a difference between data from at least two support regions from among the first plurality of support regions in which the calibration nucleic acid sample is deposited; and
applying, for the experimental set of melt curve data, the calculated correction value to data from at least one of the second plurality of support regions in which the at least one test nucleic acid sample is deposited.