Method and apparatus for detection, identification and quantification of single-and multi-analytes in affinity-based sensor arrays

1. A method of processing affinity-based array data, comprising:

(a) determining an estimate of a probe and a target interaction in an array comprising (i) analytical expression and (ii) by calibration of the array using at least one standard probe;

(b) generating a stochastic-matrix that utilizes the estimate in a Markov chain model that comprises modeling hybridization, cross-hybridization, and unbound transition probabilities between states;

(c) obtaining optical affinity-based array data using a light detection device;

(d) applying the array data to the stochastic-matrix;

(e) applying an optimization algorithm selected from the group consisting of a maximum likelihood estimation algorithm, a maximum a-posteriori criterion, a constrained least squares calculation, and any combination thereof that exploits and does not suppress non-specific interactions by considering the non-specific interactions as interference rather than noise; and

(f) outputting optimized affinity-based array data to a user, wherein the optimized affinity-based array data has an improved signal-to-noise ratio compared to the array data obtained by using the light detection device;

wherein at least one of the steps (d)-(f) is executed by using a suitably programmed computer.

2. The method of claim 1 , wherein the array is a nucleic acid array.

3. The method of claim 1 , wherein the target is a nucleic acid.

4. The method of claim 1 , wherein the probe is a nucleic acid.

5. The method of claim 1 , wherein the analytical expression comprises calculating a AG and/or a melting temperature.

6. The method of claim 1 , wherein the calibration is by measuring the interaction of one or more targets and one or more probes.

7. The method of claim 1 , wherein the probe concentration is known and the optimization method comprises a maximum a-posteriori criterion.

8. The method of claim 1 , wherein the combination comprises a constrained least squares calculation followed by a maximum likelihood estimation or a maximum a-posteriori criterion.

9. An automated microarray system of quantifying microarray data comprising:

a light detector;

a computer comprising means for converting light detected by using the light detector to data comprising a plurality of optical measured signals, the computer further comprising instructions for causing the data to be processed by the computer according to the method of claim 1 .

10. A device for use in practicing the method of claim 1 , the device comprising a signal detection element comprising a plurality of electrodes, wherein the electrodes are used to detect signal data caused by affinity-binding on an affinity binding array and a computer program medium selected from the group consisting of a removable storage device, a disk capable of installation in a disk drive, and signals on a channel, the medium tangibly embodying program instructions for quantifying the signal, the instructions operable for causing the apparatus to:

determine an estimate of a probe and a target interaction in the affinity binding array comprising (i) analytical expression and (ii) by calibration of the array using at least one standard probe;

generate a stochastic-matrix using the estimate in a Markov chain model that comprises modeling hybridization, cross-hybridization and unbound transition probabilities between states;

apply signal data to the stochastic-matrix;

apply an optimization algorithm selected from the group consisting of a maximum likelihood estimation algorithm, a maximum a-posteriori criterion, a constrained least squares calculation, and any combination thereof that exploits and does not suppress non-specific interactions by considering the non-specific interactions as interference rather than noise; and

output specific binding data that depicts substantially accurate specific binding from the signal data.

11. The device of claim 10 , wherein the device is an integrated device.

12. The device of claim 11 , wherein the device has a chip configuration.

13. A computer program, residing on a computer program medium selected from the group consisting of a removable storage device, a disk capable of installation in a disk drive, and non-transitory signals on a channel, the medium comprising instructions for causing a computer to:

determine an estimate of a probe and a target interaction in an affinity-based array comprising (i) analytical expression and (ii) by calibration of the array using at least one standard probe;

generate a stochastic-matrix that utilizes the estimate in a Markov chain model that comprises modeling hybridization, cross-hybridization, and unbound transition probabilities between states;

obtain optical image affinity-based array data using a light detection device;

apply experimental optical image data to the stochastic-matrix;

apply an optimization algorithm selected from the group consisting of a maximum likelihood estimation algorithm, a maximum a-posteriori criterion, a constrained least squares calculation, and any combination thereof that exploits and does not suppress non-specific interactions by considering the non-specific interactions as interference rather than noise; and

output specific binding data that depicts substantially accurate specific binding.

14. An apparatus comprising

a light detection device that is used to obtain microarray image data from an array;

a computer-readable storage medium tangibly embodying program instructions for quantifying microarray image data, the instructions operable for causing the apparatus to:

determine an estimate of a probe and a target interaction in an array comprising (i) analytical expression and (ii) by calibration of the array using at least one standard probe;

generate a stochastic-matrix using the estimate in a Markov chain model that comprises modeling hybridization, cross-hybridization, and unbound transition probabilities between states;

apply experimental data comprising optical signal data of the microarray image data to the stochastic-matrix;

apply an optimization algorithm selected from the group consisting of a maximum likelihood estimation algorithm, a maximum a-posteriori criterion, a constrained least squares calculation, and any combination thereof that exploits and does not suppress non-specific interactions by considering the non-specific interactions as interference rather than noise; and

output specific binding data that depicts substantially accurate specific binding from the microarray image data.