Computer-aided probability base calling for arrays of nucleic acid probes on chips

• US 5,733,729 A
• Filed: 09/14/1995
• Issued: 03/31/1998
• Est. Priority Date: 09/14/1995
• Status: Expired due to Term

• Current Assignee: Affymetrix, Inc. (Thermo Fisher Scientific Incorporated)
• Original Assignee: Affymetrix, Inc. (Thermo Fisher Scientific Incorporated)
• Inventors: Lipshutz, Robert J., Walker, Michael G.
• Primary Examiner(s): Jones, W. Gary
• Assistant Examiner(s): TRAN, PAUL
• Application Number:
  US08/528,656
• Time in Patent Office: 929 Days
• Field of Search: 935/6, 935/7.92, 935/973, 935/88, 536/23.1, 536/24.32, 364/413.01, 364/497, 382/129
• US Class Current: 435/6.11
• CPC Class Codes: C12Q 1/68 involving nucleic acids
• CPC Class Codes: C12Q 1/6837 using probe arrays or probe chips C12Q1/6874 takes precedence
• CPC Class Codes: G16B 25/00 ICT specially adapted for hybridisation; ICT specially adapted for gene or protein expression
• CPC Class Codes: Y10S 435/973 Simultaneous determination of more than one analyte

Abstract

A computer system for analyzing nucleic acid sequences is provided. The computer system is used to calculate probabilities for determining unknown bases by analyzing the fluorescence intensities of hybridized nucleic acid probes on biological chips. Additionally, information from multiple experiments is utilized to improve the accuracy of calling unknown bases.

Claims

1. In a computer system, a method of calling an unknown base in a sample nucleic acid sequence, the method comprising the steps of:

defining a set of potential base calls, said base calls including at least one of A, C, G, T(U), deletion, insertion, and a plurality of ambiguous calls;

inputting a plurality of measurements for hybridization between a plurality of probes and said sample nucleic acid sequence or nucleic acid derived from said sample nucleic acid sequence;

determining a plurality of probabilities, each of said probabilities reflecting the likelihood that one of the potential base calls is correct, said probabilities being calculated according to a distribution model, said hybridization measurements and sequences of said plurality of probes; and

calling the unknown base according to the probabilities of said potential base calls.

2. The method of claim 1, wherein the unknown base is called according to the highest probability, said highest probability indicates a confidence that the unknown base is called correctly.

3. The method of claim 1, wherein each nucleic acid probe has a different base at an interrogation position.

4. The method of claim 3, wherein each probability indicates the likelihood that the unknown base is complementary to a base at the interrogation position in the corresponding nucleic acid probe.

5. The method of claim 1, wherein a plurality of measurements for each hybridization are inputted.

6. The method of claim 5, wherein a mean of said plurality of measurements for each hybridization is determined.

7. The method of claim 5, further comprising the step of calculating a standard deviation of said plurality of measurements for each hybridization.

8. The method of claim 1, further comprising the step of comparing the highest probability to a probability threshold.

9. The method of claim 8, wherein the unknown base is called according to the nucleic acid probe associated with the highest probability if the highest probability exceeds the probability threshold.

10. The method of claim 8, further comprising the steps of:

producing a sum of the highest probability and a next highest probability;

comparing the sum to the probability threshold; and

calling the unknown base according to the nucleic acid probes associated with the highest and next highest probabilities if the sum exceeds the probability threshold.

11. The method of claim 10, wherein the sum indicates a confidence that the unknown base is called correctly if the sum exceeds the probability threshold.

12. The method of claim 1, wherein each probability is determined by a first equation:

space="preserve" listing-type="equation">Prob(X)=Prob(I.sub.x >

max(I.sub.y)).sub.yx

where X and Y are A, C, G, T(U) or deletion, insertion, or a plurality of ambiguous calls;

I is a measurement associated with a subscripted base; and

max represents a maximum of said measurements.

13. The method of claim 12, wherein the first equation is solved by a second equation:

space="preserve" listing-type="equation">Prob(I.sub.x >

max(I.sub.y))=Π

Prob(I.sub.x >

I.sub.y).sub.y≠

x

where X and Y are A, C, G, T(U), deletion, insertion, or a plurality of ambiguous calls and II represents a product of probabilities.

14. The method of claim 13, wherein the term Prob(I_x >

I_y) is solved with a third equation;

##EQU4## where X and Y are A, C, G, T(U), deletion, insertion, or a plurality of ambiguous calls and represents a standard deviation (² being a variance) of measurements for a subscripted base and represents a density equation.

15. The method of claim 14, wherein the of said third equation is:

##EQU5## and the of said third equation represents a density equation of standard normal distribution.

16. The method of claim 1, wherein the unknown base is called as being A, C, G, or T(U).

17. A computer program that calls an unknown base in a sample nucleic acid sequence, comprising:

code that receives as input a plurality of hybridization probe intensities, each of the probe intensities corresponding to a nucleic acid probe;

code that determines for each of the plurality of probe intensities a probability that the corresponding nucleic acid probe best hybridizes with the sample nucleic acid sequence; and

code that calls the unknown base according to the nucleic acid probe with the highest associated probability;

wherein the codes are stored on a tangible medium.

18. In a computer system, a method of calling an unknown base in a sample nucleic acid sequence, the method comprising the steps of:

inputting a plurality of base calls for the unknown base, each of the base calls having probability which represents a confidence that the known base is called correctly;

selecting a base call that has a highest associated probability; and

calling the unknown base according to the selected base call.

19. The method of claim 18, further comprising the step of performing a base call of the unknown base according to hybridization of nucleic acid probes with the sample nucleic acid sequence.

20. The method of claim 18, wherein the base calls are determined from a plurality of experiments.

21. A computer program that calls an unknown base in a sample nucleic acid sequence, comprising:

code that receives as input a plurality of base calls for the unknown base, each of the base calls having an associated probability which represents a confidence that the unknown base is called correctly;

selecting a base call that has a highest associated probability; and

calling the unknown base according to the selected base call;

wherein the codes are stored on a tangible medium.

22. In a computer system, a method of calling an unknown base in a sample nucleic acid sequence, the method comprising the steps of:

inputting a plurality of probabilities for each possible base for the unknown base, each of the probabilities representing a probability that the unknown base is an associated base;

producing a product of probabilities for each possible base, each product being associated with a possible base; and

calling the unknown base according to a base associated with a highest product.

23. The method of claim 22, further comprising the step of calculating the probabilities according to hybridization of nucleic acid probes with the sample nucleic acid sequence.

24. The method of claim 22, wherein the probabilities are determined from a plurality of experiments.

25. A computer program that calls an unknown base in a sample nucleic acid sequence, comprising:

code that receives as input a plurality of probabilities for each possible base for the unknown base, each of the probabilities representing a probability that the unknown base is an associated base;

code that produces a product of probabilities for each possible base, each product being associated with a possible base; and

code that calls the unknown base according to a base associated with a highest product;

wherein the codes are stored on a tangible medium.

26. In a computer system, a method of calling an unknown base in a sample nucleic acid sequence, the method comprising the steps of:

inputting a first base call for the unknown base, the first base call determined from a first nucleic acid probe that is equivalent to a portion of the sample nucleic acid sequence including the unknown base;

inputting a second base call for the unknown base, the second base call determined from a second nucleic acid probe that is complementary to a portion of the sample nucleic acid sequence including the unknown base;

selecting one of the first or second nucleic acid probes that has a base at an interrogation position which has a high probability of producing correct base calls; and

calling the unknown base according to the selected one of the first or second nucleic acid probes.

27. The method of claim 26, further comprising the step of calculating the first and second base calls according to hybridization of nucleic acid probes with the sample nucleic acid sequence.

28. The method of claim 26, wherein the base that has a high probability of producing correct base calls is A or G.

29. A computer program that calls an unknown base in a sample nucleic acid sequence, comprising:

code that receives as input first and second base calls for the unknown base, the first base call determined from a first nucleic acid probe that is equivalent to a portion of the sample nucleic acid sequence including the unknown base and the second base call determined from a second nucleic acid probe that is complementary to a portion of the sample nucleic acid sequence including the unknown base;

code that selects one of the first or second nucleic acid probes that has a base at an interrogation position which has a high probability of producing correct base calls; and

code that calls the unknown base according to the selected one of the first or second nucleic acid probes;

wherein the codes are stored on a tangible medium.

30. In a computer system, a method for determining whether a target nucleic acid hybridizes better to one of two probes comprising the steps of:

obtaining a plurality of first signals from a first hybridization between said target nucleic acid and a first probe;

obtaining a plurality of second signals from a second hybridization between said target nucleic acid and a second probe; and

calculating a probability that said first signals are stronger than said second signals according to a signal distribution model.

31. The method of claim 30, wherein said first and second signals are obtained by scanning different areas of their corresponding hybridization cells.

32. The method of claim 30, wherein said signal distribution model is the normal distribution function.

33. The method of claim 30, wherein said signal distribution model is the Student'"'"'s t distribution.

34. The method of claim 30, wherein said probability is determined by:

##EQU6## where Φ

represents a density equation of standard normal distribution;

I_x represents mean intensity of said first signals;

I_y represents mean intensity of said second signals;

σ

_x represents standard deviation of said first signals; and

σ

_y represents standard deviation of said second signals.