Claims
- 1. A method of analysis, wherein molecular interactions at one or more sensing surface areas are detected and respective response curves representing the progress of each interaction with time are produced, and wherein a resulting set of response curves is subjected to a quality assessment procedure, the method comprising the steps of:
a) selecting at least one quality-related parameter for the response curves, and for each different parameter defining at least one quality descriptor; b) computing for each response curve in the set thereof, values for the different quality descriptors; c) based on the values for the different quality descriptors, computing for each response curve a quality classification indicative of the quality of the response curve in relation to all response curves of the set; d) selecting response curves having deviating quality classifications; and e) subjecting the selected response curves to a validation procedure to determine whether a response curve or curves are to be rejected or not.
- 2. The method of claim 1 wherein steps c) to e) are repeated until no more response curves are rejected.
- 3. The method of claim 1 wherein step b) of claim 1 further comprises transforming the quality descriptor values for each response curve to a quality descriptor vector.
- 4. The method of claim 3 wherein a quality descriptor matrix is created from the quality descriptor vectors.
- 5. The method of claim 1 wherein the descriptor values are normalized.
- 6. The method of claim 1 wherein computing a quality classification in step c) of claim 1 comprises determining for each quality descriptor vector the difference between the vector and the rest of the quality descriptor vectors in the set of response curves.
- 7. The method of claim 6 wherein the determination of the differences between the vectors comprises determining a statistical measure of the distance from each quality descriptor vector to the rest of the quality descriptor vectors as a group.
- 8. The method of claim 6 wherein a difference vector is created from the computed differences.
- 9. The method of claim 6 wherein step d) of claim 1 comprises sorting the quality vectors for the response curves in order of computed difference, and selecting a predetermined fraction of the response curves corresponding to those having the largest differences.
- 10. The method of claim 1 wherein computing a quality classification in step c) of claim 1 comprises classifying the sensorgrams in groups of similar quality.
- 11. The method of claim 1 wherein the quality classification is obtained by a neural network.
- 12. The method of claim 1 wherein the quality classification is obtained by an expert system.
- 13. The method of claim 1 wherein the validation procedure in step e) of claim 1 comprises subjecting selected response curves to an ocular inspection.
- 14. The method of claim 1 wherein the validation procedure in step e) of claim 1 comprises determining which quality descriptor or descriptors that caused the quality classification of a sensorgram.
- 15. The method of claim 1 wherein the validation procedure in step e) of claim 1 comprises identifying a time period or periods in the production of the sensorgrams to which a cluster of selected sensorgrams may be related.
- 16. The method of claim 1 wherein the at least one quality-related parameter on which the quality descriptor or descriptors are based comprises at least one parameter selected from an odd curve shape, carry-over between measurements, binding to a reference surface, and dissociation to a value below zero.
- 17. The method of claim 16 wherein the odd curve shape is selected from sloping baseline, spikes, oscillations and jumps.
- 18. The method of claim 2 wherein steps c) to e) of claim 1 are repeated with at least one different descriptor.
- 19. The method of claim 2 wherein steps c) to e) of claim 1 are repeated with a different quality classification method.
- 20. The method of claim 1 wherein a biosensor is used.
- 21. The method of claim 20 wherein the biosensor is based on evanescent wave sensing, particularly surface plasmon resonance.
- 22. A computer-implemented method of analysis, wherein molecular interactions at one or more sensing surface areas are detected and respective response curves representing the progress of each interaction with time are produced, and wherein a resulting set of response curves is subjected to a quality assessment procedure, the method comprising the steps of:
a) selecting at least one quality-related parameter for the response curves, and for each different parameter defining at least one quality descriptor; b) computing for each response curve in the set thereof, values for the different quality descriptors; c) based on the values for the different quality descriptors, computing for each response curve a quality classification indicative of the quality of the response curve in relation to all response curves of the set; d) selecting response curves having deviating quality classifications; and e) subjecting the selected response curves to a validation procedure to determine whether a response curve or curves are to be rejected or not.
- 23. An analytical system for detecting molecular binding interactions, comprising:
(a) a sensor device comprising at least one sensing surface, detection means for detecting molecular interactions at the at least one sensing surface, and means for producing response curves representing the progress of each interaction with time; and (b) data processing means for classifying the response curves with regard to quality.
- 24. The analytical system of claim 23 wherein the data processing means perform the steps of:
a) selecting at least one quality-related parameter for the response curves, and for each different parameter defining at least one quality descriptor; b) computing for each response curve in the set thereof, values for the different quality descriptors; c) based on the values for the different quality descriptors, computing for each response curve a quality classification indicative of the quality of the response curve in relation to all response curves of the set; d) selecting response curves having deviating quality classifications; and e) subjecting the selected response curves to a validation procedure to determine whether a response curve or curves are to be rejected or not.
- 25. A computer program comprising program code means for performing a procedure for assessing the quality of response curves resulting from analysis of molecular interactions detected at one or more sensing surface areas, which response curves represent the progress of each interaction with time, when the program is run on a computer, which procedure comprises the steps of:
a) selecting at least one quality-related parameter for the response curves, and for each different parameter defining at least one quality descriptor; b) computing for each response curve in the set thereof, values for the different quality descriptors; c) based on the values for the different quality descriptors, computing for each response curve a quality classification indicative of the quality of the response curve in relation to all response curves of the set; d) selecting response curves having deviating quality classifications; and e) subjecting the selected response curves to a validation procedure to determine whether a response curve or curves are to be rejected or not.
- 26. A computer program product comprising program code means for performing a procedure for assessing the quality of response curves resulting from analysis of molecular interactions detected at one or more sensing surface areas, which response curves represent the progress of each interaction with time, when the program is run on a computer, which procedure comprises the steps of:
a) selecting at least one quality-related parameter for the response curves, and for each different parameter defining at least one quality descriptor; b) computing for each response curve in the set thereof, values for the different quality descriptors; c) based on the values for the different quality descriptors, computing for each response curve a quality classification indicative of the quality of the response curve in relation to all response curves of the set; d) selecting response curves having deviating quality classifications; and e) subjecting the selected response curves to a validation procedure to determine whether a response curve or curves are to be rejected or not.
- 27. The computer program product according to claim 26 wherein the program code means are stored on a computer readable medium.
- 28. The computer program product according to claim 26 wherein the program code means are stored in a computer memory.
- 29. The computer program product according to claim 26 wherein the program code means are carried on a carrier selected from an electrical signal and an optical signal.
- 30. A computer system containing a program for performing a procedure for assessing the quality of response curves resulting from analysis of molecular interactions detected at one or more sensing surface areas, which response curves represent the progress of each interaction with time, when the program is run on a computer, which procedure comprises the steps of:
a) selecting at least one quality-related parameter for the response curves, and for each different parameter defining at least one quality descriptor; b) computing for each response curve in the set thereof, values for the different quality descriptors; c) based on the values for the different quality descriptors, computing for each response curve a quality classification indicative of the quality of the response curve in relation to all response curves of the set; d) selecting response curves having deviating quality classifications; and e) subjecting the selected response curves to a validation procedure to determine whether a response curve or curves are to be rejected or not.
Priority Claims (1)
Number |
Date |
Country |
Kind |
0200949-6 |
Mar 2002 |
SE |
|
CROSS-REFERENCE TO RELATED APPLICATION
[0001] This application claims the benefit of U.S. Provisional Patent Application No. 60/367,806 filed Mar. 27, 2002 and Swedish Patent Application No. 0200949-6 filed Mar. 27, 2002, both of which are incorporated herein by reference in their entireties.
Provisional Applications (1)
|
Number |
Date |
Country |
|
60367806 |
Mar 2002 |
US |