Computational methods and systems for multidimensional analysis

Abstract

A method for analyzing data obtained from at least one sample in a separation system ( 10, 50, 60 ) that has a capability for separating components of a sample containing more than one component as a function of at least two different variables comprising obtaining data representative of the at least one sample from the system, the data being expressed as a function of the two variables; forming a data stack ( 70, 74, 78, 82, 84 ) having successive levels, each level containing successive data representative of the at least one sample; forming a data array (R) representative of a compilation of all of the data in the data stack; and separating the data array into a series of matrixes. A chemical analysis system that operates in accordance with the method, and a medium having computer readable program code for causing the system to perform the method.

1 . A method for analyzing data obtained from a sample in a separation system that has a capability for separating components of a sample containing more than one component, said method comprising:
 separating said sample with respect to at least a first variable to form a separated sample;   separating said separated sample with respect to at least a second variable to form a further separated sample;   obtaining data representative of said further separated sample from a multi-channel analyzer, said data being expressed as a function of three variables;   forming a data stack having successive levels, each level containing data from one channel of said multi-channel analyzer;   forming a data array representative of a compilation of all of the data in said data stack; and   separating said data array into a series of matrixes or arrays, said matrixes or arrays being:   a concentration data array representative of concentration of each component in said sample on its super-diagonal;   a first profile of each component as a function of a first variable;   a second profile of each component as a function of a second variable; and   a third profile of each component as a function of a third variable.   
   
   
       2 . The method of  claim 1 , wherein said first profile, said second profile, and said third profile are representative of profiles of substantially pure components. 
   
   
       3 . The method of  claim 1 , further comprising performing qualitative analysis using at least one of said first profile, said second profile, and said third profile. 
   
   
       4 . The method of  claim 1 , further comprising standardizing data representative of a sample by performing a data matrix multiplication of such data into the product of a first standardization matrix, the data itself, and a second standardization matrix, to form a standardized data matrix. 
   
   
       5 . The method of  claim 4 , wherein terms in said first standardization matrix and said second standardization matrix have values that cause said data to be represented at positions with respect to two of said three variables, which are different in said standardized data matrix from those in said data array. 
   
   
       6 . The method of  claim 5 , wherein said first standardization matrix shifts said data with respect to one of said two variables, and said second standardization matrix shifts said data with respect to the other of said two variables. 
   
   
       7 . The method of  claim 5 , wherein terms in said first standardization matrix and said second standardization matrix have values that serve to standardize distribution shapes of the data with respect to said two variables, respectively. 
   
   
       8 . The method of  claim 4 , wherein terms in said first standardization matrix and said second standardization matrix are determined by:
 applying a sample having known components to said apparatus; and   selecting terms for said first standardization matrix and said second standardization matrix which cause data produced by said known components to be positioned properly with respect to the two variables.   
   
   
       9 . The method of  claim 8 , wherein said terms are determined by selecting terms which produce a smallest error in position of said data with respect to the two variables, in said standardized data matrix. 
   
   
       10 . The method of  claim 9 , wherein the terms of said first standardization matrix and said second standardization matrix are computed for a single channel. 
   
   
       11 . The method of  claim 10 , wherein terms of said first standardization matrix and said second standardization matrix are computed so as to produce a smallest error for the channel. 
   
   
       12 . The method of  claim 4 , wherein at least one of the first and second standardization matrices can be simplified to be either a diagonal matrix or an identity matrix. 
   
   
       13 . The method of  claim 4 , wherein the terms in said first standardization matrix and said second standardization matrix are based on parameterized known functional dependence of said terms on said variables. 
   
   
       14 . The method of  claim 4 , wherein values of terms in said first standardization matrix and in said second standardization matrix are determined by solving data array R: 
     
       
         
         
             
             
         
       
     
     where Q (m×k) contains pure profiles of all k components with respect to the first variable, W (n×k) contains pure profiles with respect to the second variable for the components, C (p×k) contains pure profiles of these components with respect to the multichannel analyzer or the third variable, I (k×k×k) is a new data array with scalars on its super-diagonal as the only nonzero elements representing the concentrations of all said k components, and E (m×n×p) is a residual data array. 
   
   
       15 . The method of  claim 1 , wherein one of said separation apparatus is a one-dimensional electrophoresis separation system. 
   
   
       16 . The method of  claim 15 , wherein said variable is one of isoelectric point and molecular weight. 
   
   
       17 . The method of  claim 1 , wherein said two separation variables are a result of any combination, in no particular sequence, and including self-combination, of chromatographic separation, capillary electrophoresis separation, gel-based separation, affinity separation and antibody separation 
   
   
       18 . The method of  claim 1 , wherein one of the three variables is mass associated with the mass axis of a mass spectrometer. 
   
   
       19 . The method of  claim 18 , wherein said apparatus further comprises at least one chromatography system for providing said separated samples to said mass spectrometer, retention time being at least one of the variables. 
   
   
       20 . The method of  claim 18 , wherein said apparatus further comprises at least one electrophoresis separation system for providing said separated samples to said mass spectrometer, migration characteristics of said sample being at least one of the variables. 
   
   
       21 . The method of  claim 18 , wherein said data is continuum mass spectral data. 
   
   
       22 . The method of  claim 18 , wherein said data is used without centroiding. 
   
   
       23 . The method of  claim 18 , further comprising correcting said data for time skew. 
   
   
       24 . The method of  claim 18 , further comprising performing a calibration of said data with respect to mass and spectral peak shapes. 
   
   
       25 . The method of  claim 18 , wherein said apparatus comprises a protein chip having a plurality of protein affinity regions, location of a region being one of said three variables. 
   
   
       26 . The method of  claim 1 , wherein said multichannel analyzer is based on one of light absorption, light emission, light reflection, light transmission, light scattering, refractive index, electrochemistry, conductivity, radioactivity, or any combination thereof. 
   
   
       27 . The method of  claim 26 , wherein the components in said sample are bound to at least one of fluorescence tags, isotope tags, stains, affinity tags, or antibody tags. 
   
   
       28 . The method of  claim 1 , wherein said apparatus comprises a two-dimensional electrophoresis separation system. 
   
   
       29 . The method of  claim 28 , wherein a first of said at least one variable is isoelectric point and a second of said at least one variable is molecular weight. 
   
   
       30 . A computer readable medium having thereon computer readable code for use with a chemical analysis system having a data analysis portion for analyzing data obtained from a sample, said chemical analysis system having a separation portion that has a capability for separating components of a sample containing more than one component as a function of at least one variable, said computer readable code being for causing the computer to perform a method comprising:
 separating said sample with respect to at least a first variable to form a separated sample;   separating said separated sample with respect to at least a second variable to form a further separated sample;   obtaining data representative of said further separated sample from a multi-channel analyzer, said data being expressed as a function of three variables;   forming a data stack having successive levels, each level containing data from one channel of said multi-channel analyzer;   forming a data array representative of a compilation of all of the data in said data stack; and   separating said data array into a series of matrixes or arrays, said matrixes or arrays being:   a concentration data array representative of concentration of each component in said sample on its super-diagonal;   a first profile of each component as a function of a first variable;   a second profile of each component as a function of a second variable; and   a third profile of each component as a function of a third variable.   
   
   
       31 . A chemical analysis system for analyzing data obtained from a sample, said system having a separation system that has a capability for separating components of a sample containing more than one component as a function of at least one variable, said system having apparatus for performing a method comprising:
 separating said sample with respect to at least a first variable to form a separated sample;   separating said separated sample with respect to at least a second variable to form a further separated sample;   obtaining data representative of said further separated sample from a multi-channel analyzer, said data being expressed as a function of three variables;   forming a data stack having successive levels, each level containing data from one channel of said multi-channel analyzer;   forming a data array representative of a compilation of all of the data in said data stack; and   separating said data array into a series of matrixes or arrays, said matrixes or arrays being:   a concentration data array representative of concentration of each component in said sample on its super-diagonal;   a first profile of each component as a function of a first variable;   a second profile of each component as a function of a second variable; and   a third profile of each component as a function of a third variable.