Metabolizer

User Tools

Site Tools

Trace:
prediction

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
prediction [2016/12/23 00:53] – ccubukprediction [2017/05/24 14:33] (current) – external edit 127.0.0.1
Line 1: Line 1:
 ====== Prediction ====== ====== Prediction ======
 +
 +===== Train =====
  
 The aim of the //Prediction// tool is to take advantage of the module activities to distinguish between phenotypes.  The aim of the //Prediction// tool is to take advantage of the module activities to distinguish between phenotypes. 
Line 5: Line 7:
 Metabolizer Prediction uses the module activity values to compute a RF or SVM prediction model with cross-validation. Moreover, previously obtained models can be used to predict the phenotype of new samples. Metabolizer Prediction uses the module activity values to compute a RF or SVM prediction model with cross-validation. Moreover, previously obtained models can be used to predict the phenotype of new samples.
  
-RF: Random Forest +RF: Random Forest, SVM: Support Vector Machines
-SVM: Support Vector Machines+
  
 The tool can be accessed from the main menu bar, by clicking on the //Prediction// button, see [[workflow|Workflow]] for further information. The tool can be accessed from the main menu bar, by clicking on the //Prediction// button, see [[workflow|Workflow]] for further information.
Line 24: Line 25:
 === OUTPUT === === OUTPUT ===
  
-The results page of the Prediction tool includes different output results. You can download any table or image showed in the results page by clicking on the name right before it. You can also download the pathway and function matrices by clicking on //Path values//.+The results page of the Prediction tool includes different output results. You can download any table or image showed in the results page by clicking on the name right before it. You can also download calculated module activity values and CV statistics of trained model by clicking on //**Module activity values**// and //**Statistics**// respectively. CV results also represented by area under the Receiver Operating characteristic (ROC) curve. 
 + 
 +CV: Cross Validation
  
 The results are divided in different panels: The results are divided in different panels:
  
   * [[Input Parameters Prediction|Input Parameters]]   * [[Input Parameters Prediction|Input Parameters]]
-  * **Path values**: You can download the matrix of path values by clicking on Path values.+  * [[path values prediction | Path values]]
   * [[Prediction Model|Prediction Model]]   * [[Prediction Model|Prediction Model]]
 +
 +===== Test =====
 +
 +  * [[Input Parameters Prediction Test|Input Parameters]]
 +  * [[path values prediction test | Path values]]
 +  * [[Prediction Results |Prediction Results]]
prediction.1482454430.txt.gz · Last modified: 2017/05/24 14:33 (external edit)