DNA-based analog neural networks

• US 20030022164A1
• Filed: 12/21/2000
• Published: 01/30/2003
• Est. Priority Date: 08/06/1998
• Status: Active Grant

• Current Assignee: WSOU Investments, LLC (WSOU Holdings, LLC)
• Original Assignee: Lucent Technologies, Inc. (Nokia Corporation)
• Inventors: Mills, Allen P. Jr.
• Granted Patent: US 7,297,479 B2
• Time in Patent Office: Days
• US Class Current: 435/6
• CPC Class Codes: B82Y 10/00 Nanotechnology for information processing, storage or transmission, e.g. quantum computing or single electron logic
• CPC Class Codes: G06N 3/02 Neural networks
• CPC Class Codes: G06N 3/123 DNA computing

Abstract

This invention is an oligomer-based analog neural network (ANN) comprising weight and saturation oligomers, the concentrations of which are selected such that activation of the ANN by a set of input oligomers generates a set of output oligomers, the sequences and relative concentrations of which are dependent on the sequences and relative concentrations of the input oligomers. The invention further includes methods for using such an ANN for solving any problems amenable to solution by a trained neural network. A preferred embodiment of the claimed invention is a DNA-based ANN that accepts cDNA molecules as inputs and analyzes the gene expression profile of the cells from which the cDNA is derived. The DNA-based ANN is typically trained with a computer to identify the weights giving accurate mapping of the inputs to the outputs; and the concentrations of weight oligomers of the DNA-based ANN are then selected accordingly.

Claims

1. An aqueous solution for use in performing algebraic operations, comprising:

a plurality of types of oligomers comprising a plurality of first subunit sequences and a plurality of second subunit sequences;

wherein each oligomer comprises first and second serially connected subunit sequences, and is a nucleic acid or an analog of a nucleic acid; and

wherein the first subunit sequences hybridize to nucleic acid molecules produced by a reference cell, and the second subunit sequences are configured so that they do not hybridize to nucleic acid molecules produced by said cell.

2. The solution of claim 1, wherein the first and second subunit sequences are single-stranded and double-stranded, respectively.

3. The solution of claim 1, wherein the second sequences are not complementary to nucleic acid molecules produced by the cell.

4. The solution of claim 2, wherein the plurality of types includes oligomers whose second sequences are complementary to each other.

5. A method of performing expression profiling of a cell, comprising:

providing a solution with a plurality of types of oligomers comprising a plurality of first subunit sequences and a plurality of second subunit sequences;

wherein each oligomer comprises first and second serially connected subunit sequences, and is a nucleic acid or an analog of a nucleic acid; and

wherein the first subunit sequences hybridize to nucleic acid molecules produced by a reference cell, and the second subunit sequences are configured so that they do not hybridize to nucleic acid molecules produced by said cell;

and mixing an assay of nucleic acid oligomers representative of oligomers of a test cell in the solution.

6. The method of claim 5, further comprising:

determining a condition of the test cell from concentrations of second sequences of the oligomers that hybridize as a result of the mixing.

7. The method of claim 5, wherein the first and second sequences are single-stranded and double-stranded, respectively.

8. The method of claim 5, wherein the second sequences are not complementary to sequences of nucleic acids produced by the test cell.

9. The method of claim 7, wherein the mixing includes providing conditions in which oligomers of the test cell hybridize to single-stranded sequences of a portion of the oligomers of the solution.

10. The method of claim 5, further comprising:

extracting mRNA from the test cell; and

forming the assay by amplifying the extracted mRNA.

11. The method of claim 10, wherein the amplifying includes forming a DNA copy of at least a portion of the extracted mRNA and producing further copies of the DNA copy.

12. The method of claim 5, further comprising:

deciding whether the test and reference cells have similar expression profiles by identifying second sequences of the weight oligomers that hybridize in response to the mixing.

13. A method of performing expression profiling, comprising:

extracting RNA from a test cell;

forming an assay by amplifying the extracted mRNA; and

mixing the assay into a solution that includes a a plurality of types of oligomers comprising a plurality of first subunit sequences and a plurality of second subunit sequences;

wherein each oligomer comprises first and second serially connected subunit sequences, and is a nucleic acid or an analog of a nucleic acid; and

wherein the first subunit sequences hybridize to nucleic acid molecules produced by a cell, and the second subunit sequences are configured so that they do not hybridize to nucleic acid molecules produced by said cell.

14. The method of claim 13, further comprising:

identifying the second sequences of the weight oligomers whose first sequences hybridize in response to being mixed with the assay.

15. The method of claim 13, further comprising:

determining concentrations of second sequences associated with the portion of the oligomers that hybridize to nucleic acid molecules in the assay in response to the mixing; and

associating a type to the test cell in response to the concentrations of the separated second sequences.

16. The method of claim 13, further comprising:

determining a type of the test cell based on concentrations of second sequences associated with the portion of the oligomers that hybridize to nucleic acid molecules in the assay in response to the mixing.

17. The method of claim 16, wherein the type of the test cell is one of diseased and healthy.

18. A method for obtaining a set of output oligomers representing the result of at least one algebraic operation, the sequences and relative concentrations of which are dependent on the sequences and relative concentrations of a set of input oligomers, comprising

combining said set of input oligomers with at least one set of weight oligomers, and

determining the result of an algebraic operation based on the concentrations of output oligomers produced by the combining;

said input, weight, and output oligomers independently being nucleic acids or nucleic acid analogs.

19. The method of claim 18 wherein said weight oligomers comprise a plurality of types of oligomers, each oligomer comprising at least one single-stranded receptor oligomer serially connected to at least one signaling oligomer.

20. The method of claim 19 wherein each weight oligomer comprises a receptor oligomer strand that is connected to one strand of a double-stranded signaling oligomer.

21. The method of claim 18, wherein the combining comprises hybridizing an input oligomer to a receptor oligomer portion of a weight oligomer.

22. The method of claim 18, wherein the determining comprises obtaining signal oligomers from weight oligomers that are hybridized to input oligomers.

23. The method of claim 18, wherein the determining comprises obtaining signal oligomers that are complementary to each other.

24. The method of claim 22, wherein the determining further comprises hybridizing the signal oligomers to at least one set of saturation oligomers to obtain a set of saturated signal oligomers.

25. The method of claim 24 wherein the set of saturated signal oligomers are the output oligomers.