Methods for Analyzing High Dimensional Data for Classifying, Diagnosing, Prognosticating, and/or Predicting Diseases and Other Biological States

Abstract

A method of diagnosing, predicting, or prognosticating about a disease that includes obtaining experimental data, wherein the experimental data is high dimensional data, filtering the data, reducing the dimensionality of the data through use of one or more methods, training a supervised pattern recognition method, ranking individual data points from the data, wherein the ranking is dependent on the outcome of the supervised pattern recognition method, choosing multiple data points from the data, wherein the choice is based on the relative ranking of the individual data points, and using the multiple data points to determine if an unknown set of experimental data indicates a diseased condition, a predilection for a diseased condition, or a prognosis about a diseased condition.

1 .- 10 . (canceled) 
     
     
         11 . A method of determining whether a subject has a cancer selected from the group consisting of neuroblastoma, rhabdomyosarcoma, non-Hodgkins lymphoma, and Ewings tumor comprising:
 (a) detecting a gene expression level of at least 10 genes selected from the group consisting of SEQ ID NO: 1 to SEQ ID NO:96 in a biological sample from a subject whose cancer is unknown to obtain high dimensional experimental data;   (b) detecting the presence of a first set of multiple data points in the high dimensional experimental data, wherein the first set of multiple data points is predictive for distinguishing the presence of neuroblastoma, rhabdomyosarcoma, non-Hodgkins lymphoma, and Ewings tumor from one another as determined by a trained supervised pattern recognition method; and   (c) determining whether a subject whose cancer is unknown has neuroblastoma, rhabdomyosarcoma, non-Hodgkins lymphoma, or Ewings tumor by inputting the first set of multiple data points into one or more probability distribution models obtained by the trained supervised pattern recognition method and obtaining a determination of the probability that the sample is indicative of a cancer selected from the group consisting of neuroblastoma, neuroblastoma, rhabdomyosarcoma, non-Hodgkins lymphoma, and Ewings tumor.   
     
     
         12 . The method of  claim 11 , wherein processing the biological sample comprises isolating nucleic acids or proteins from the biological sample and detecting the nucleic acids or proteins from the sample to determine gene expression levels or protein expression levels. 
     
     
         13 . The method of  claim 12 , wherein said gene expression data is obtained by using a cDNA or an oligonucleotide microarray. 
     
     
         14 . The method of  claim 11 , wherein the at least 10 genes comprise SEQ ID NO:71. 
     
     
         15 . The method of  claim 11  wherein the at least 10 genes comprises SEQ ID NO:5, SEQ ID NO:20, SEQ ID NO:21, SEQ ID NO:26, SEQ ID NO:28, SEQ ID NO:43, SEQ ID NO:59, SEQ ID NO:72, SEQ ID NO:73, and SEQ ID NO:77. 
     
     
         16 . The method of  claim 11 , wherein said first set of multiple data points comprise at least 96 individual data points corresponding to each of the genes of SEQ ID NO:1 to SEQ ID NO:96. 
     
     
         17 . A computer-based method comprising:
 (a) obtaining high dimensional experimental data from a biological sample from a subject whose type of cancer is unknown, wherein the high dimensional data is obtained from gene expression levels of at least 10 genes selected from the group consisting of SEQ ID NO: 1 to SEQ ID NO:96, and providing the high dimensional data to a receiver module;   (b) detecting the presence of a first set of multiple data points in the high dimensional experimental data, wherein the first set of multiple data points is predictive for distinguishing a cancer selected from the group consisting of neuroblastoma, rhabdomyosarcoma, non-Hodgkins lymphoma, and Ewings tumor as determined by a trained supervised pattern recognition program, using a diagnostic module; and   (c) determining whether a subject whose cancer is unknown has neuroblastoma, rhabdomyosarcoma, non-Hodgkins lymphoma, or Ewings tumor by inputting the first set of multiple data points into one or more probability distribution models obtained by the trained supervised pattern recognition method and obtaining a determination of the probability that the sample is indicative of a cancer selected from the group consisting of neuroblastoma, neuroblastoma, rhabdomyosarcoma, non-Hodgkins lymphoma, and Ewings tumor.   
     
     
         18 . The method of  claim 17 , wherein obtaining high dimensional experimental data from the biological sample comprises isolating nucleic acids or proteins from each biological sample and detecting the nucleic acids or proteins from each sample to determine gene expression levels or protein expression levels. 
     
     
         19 . A computer readable storage medium comprising:
 a receiver module for receiving data representing experimental gene expression data obtained from a biological sample from a subject whose type of cancer is unknown, the gene expression data comprising gene expression levels of at least 10 genes selected from the group consisting of SEQ ID NO: 1 to SEQ ID NO:96; and   a diagnostic module encoded to diagnose the presence of a cancer selected from the group neuroblastoma, rhabdomyosarcoma, non-Hodgkins lymphoma, and Ewings tumor by detecting the presence of a first set of multiple data points in gene expression data obtained from a biological sample from the subject, wherein the first set of multiple data points is predictive for the type of cancer selected from the group consisting of neuroblastoma, rhabdomyosarcoma, non-Hodgkins lymphoma, and Ewings tumor.