Rapid method of pattern recognition, machine learning, and automated genotype classification through correlation analysis of dynamic signals

Abstract

The present invention relates to methods and systems for the analysis of the dissociation behavior of nucleic acids and the identification of determining whether a genotype is present in a biological sample. The present invention includes methods and systems for determining whether a genotype is present in a biological sample, through generating a dynamic profile an unknown genotype, correlating the dynamic profile to an average profile for a known genotype to generate a correlation value, and determining whether the correlation value falls within an acceptable threshold to determine if the unknown genotype is the known genotype. The present invention also relates to methods and systems for generating a training set to allow a machine to recognize a known genotype from within a class of known genotypes through generating dynamic profiles of a known genotype, averaging the dynamic profiles to generate an average profile for the genotype, and correlating the dynamic profiles of the genotype with an average profile for each known genotype in a class of genotypes to generate a correlation vector. The training set generated by these methods and systems may be used to assist in identification of unknown genotypes.

1 . A method of determining the identity of the genotype of a nucleic acid present in a biological sample, said method comprising the steps of:
 (a) generating a dynamic profile of an unknown genotype contained in the biological sample, wherein the dynamic profile comprises measurements of a signal representing a physical change of a nucleic acid containing the unknown genotype relative to an independent variable;   (b) correlating the dynamic profile of the unknown genotype with an average dynamic profile of each known genotype in a class of known genotypes to generate a correlation vector, wherein the average dynamic profile of each known genotype comprises average measurements of a signal representing the physical change of a nucleic acid containing the known genotype relative to the independent variable and wherein the correlation vector comprises correlation coefficients between the dynamic profile of the unknown genotype and the average dynamic profile for each known genotype in the class of known genotypes; and   (c) determining whether the correlation vector or a transformation thereof falls within an acceptable range to classify the unknown genotype as one of the known genotypes in the class of known genotypes, whereby the identity of the genotype of the nucleic acid in the biological sample is determined.   
     
     
         2 - 53 . (canceled)