Identifying associations using graphical models

Abstract

Computer-executable instructions for identifying associations are described herein. By way of example, a method for facilitating developing a treatment can include employing computer-executable instructions stored on one or more computer-readable media to determine correlations and utilizing at least some of the determined correlations to develop a treatment.

1 . A method for facilitating developing a treatment, comprising:
 employing computer-executable instructions stored on one or more computer-readable media to determine correlations, the computer-executable instructions facilitating:
 receiving data relating to one or more predictor variables and a target variable across a set of instances and data relating to a phylogenetic tree; 
 generating a set of predictions based on evolution of the target variable along the phylogenetic tree; 
 generating another set of predictions based on evolution of the target variable along the phylogenetic tree until the tips of the phylogenetic tree and based on the one or more predictor variables at the tips of the phylogenetic tree; and 
 comparing the sets of predictions to determine whether the one or more predictor variables correlate with the target variable; and 
   utilizing at least some of the determined correlations to develop a treatment.   
   
   
       2 . The method of  claim 1 , the computer-executable instructions further configured to perform the steps of receiving, generating and comparing according to a set of target variables. 
   
   
       3 . The method of  claim 1 , wherein utilizing the at least some of the determined correlations to develop the treatment comprises utilizing the at least some of the determined correlations to develop a vaccine. 
   
   
       4 . The method of  claim 3 , wherein the target variable is an HIV sequence and the vaccine is an AIDS vaccine. 
   
   
       5 . The method of  claim 1 , wherein the target variable is a phenotype and at least one of the one or more predictor variables is a genotype. 
   
   
       6 . The method of  claim 1 , wherein the target variable is a genotype and at least one of the one or more predictor variables is a phenotype. 
   
   
       7 . The method of  claim 5 , wherein utilizing the at least some of the determined correlations to develop the treatment comprises utilizing the at least some of the determined correlations to develop a genetic therapy. 
   
   
       8 . The method of  claim 6 , wherein utilizing the at least some of the determined correlations to develop the treatment comprises utilizing the at least some of the determined correlations to develop a genetic therapy. 
   
   
       9 . The method of  claim 1 , wherein utilizing the at least some of the determined correlations to develop the treatment comprises utilizing the at least some of the determined correlations to develop a protein based treatment. 
   
   
       10 . The method of  claim 1 , wherein utilizing the at least some of the determined correlations to develop the treatment comprises utilizing the at least some of the determined correlations to develop a molecule that up regulates a gene product. 
   
   
       11 . The method of  claim 1 , wherein utilizing the at least some of the determined correlations to develop the treatment comprises utilizing the at least some of the determined correlations to develop a molecule that down regulates a gene product. 
   
   
       12 . A method for facilitating developing a treatment, comprising:
 employing computer-executable instructions stored on one or more computer-readable media to determine correlations, the computer-executable instructions facilitating:
 receiving data relating to one or more predictor variables and a target variable across a set of instances and data relating to a phylogenetic tree; 
 generating a set of predictions based on independent coevolution of the one or more predictor variables and the target variable along the phylogenetic tree; 
 generating another set of predictions based on evolution of the one or more predictor variables along the phylogenetic tree and evolution of the target variable along the phylogenetic tree dependent upon the one or more predictor variables; 
 comparing the sets of predictions to determine whether the one or more predictor variables correlate with the target variable; and 
   utilizing at least some of the determined correlations to develop a treatment.   
   
   
       13 . The method of  claim 12 , wherein utilizing the at least some of the determined correlations to develop the treatment comprises utilizing the at least some of the determined correlations to develop a vaccine. 
   
   
       14 . The method of  claim 12 , wherein the target variable is a phenotype and at least one of the one or more predictor variables is a genotype. 
   
   
       15 . The method of  claim 12 , wherein the target variable is a genotype and at least one of the one or more predictor variables is a phenotype. 
   
   
       16 . The method of  claim 12 , wherein utilizing the at least some of the determined correlations to develop the treatment comprises utilizing the at least some of the determined correlations to develop a genetic therapy. 
   
   
       17 . The method of  claim 12 , wherein utilizing the at least some of the determined correlations to develop the treatment comprises utilizing the at least some of the determined correlations to develop a protein based treatment. 
   
   
       18 . The method of  claim 12 , wherein utilizing the at least some of the determined correlations to develop the treatment comprises utilizing the at least some of the determined correlations to develop a molecule that regulates a gene product. 
   
   
       19 . Computer-executable instructions for performing a method of simultaneously inferring a phylogenetic tree and identifying associations among predictor and target variables, the computer-executable instructions stored on one or more computer-readable media, the method comprising:
 receiving data relating to one or more predictor variables and a target variable across a set of instances;   learning a phylogenetic tree assuming no associations between the one or more predictor variables and the target variable;   identifying associations between the one or more predictor variables and the target variable using the learned phylogenetic tree;   relearning the phylogenetic tree according to the identified associations; and   repeating identifying the associations and relearning the phylogenetic tree until the identified associations are substantially stable.   
   
   
       20 . The computer-executable instructions of  claim 19 , wherein identifying associations between the one or more predictor variables and the target variable comprises using a Bayesian cutoff for relevance and/or a frequentist cutoff for relevance.