enviPath Wiki

User Tools

Site Tools

Trace: • pathways
relative_reasoning

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
relative_reasoning [2024/03/19 08:48] – [Contents] shankarrelative_reasoning [2024/03/27 09:24] (current) – anguerasempere
Line 2: Line 2:
  
  
-From enviPath Wiki +To limit the combinatorial explosion in [[pathways|pathway]] prediction, enviPath allows to learn and apply relative reasoning models. Rule-based relative reasoning models are available, as well as multi-label machine learning algorithms. If employed during pathway prediction, models predict probabilities for each applied [[rules|rule]] and the corresponding educt and product compound, and can thus be used to [[settings|truncate]] the predicted pathway.
-Jump to navigation Jump to search +
-To limit the combinatorial explosion in pathway prediction, enviPath allows to learn and apply relative reasoning models. Rule-based relative reasoning models are available, as well as multi-label machine learning algorithms. If employed during pathway prediction, models predict probabilities for each applied rule and the corresponding educt and product compound, and can thus be used to truncate the predicted pathway. +
- +
-===== Contents ===== +
- +
- +
-  * 1 Standard models +
-  * 2 Creating models +
-  * 3 Applying relative reasoning +
-  * 4 Model types +
-      -4.1 Rule-Based Relative reasoning +
-      -4.2 Machine Learning-Based Relative Reasoning+
  
 ===== Standard models ===== ===== Standard models =====
Line 24: Line 12:
  
  
-Moreover, the user can train own such models on any collection of packages. Hence, new models can reflect exactly the conditions of the selected packages, and predicted probabilities are automatically adapted to the type of data used as input in the training process. In more detail, relative reasoning models will be trained on all compounds, rules and reactions that are available within the selected package/s.+Moreover, the user can train own such models on any collection of packages. Hence, new models can reflect exactly the conditions of the selected packages, and predicted probabilities are automatically adapted to the type of data used as input in the training process. In more detail, relative reasoning models will be trained on all [[compounds|compounds]], [[rules|rules]] and [[reactions|reactions]] that are available within the selected package/s.
  
 ===== Applying relative reasoning ===== ===== Applying relative reasoning =====
  
  
-Relative reasoning can be applied for pathway prediction by selecting/creating a dedicated setting when initializing the pathway prediction. Within each setting the relative reasoning model and a probability threshold can be selected (all edges with a predicted probability below this threshold will be pruned).+Relative reasoning can be applied for [[pathways|pathway]] prediction by selecting/creating a dedicated [[settings|setting]] when initializing the pathway prediction. Within each setting the relative reasoning model and a probability threshold can be selected (all edges with a predicted probability below this threshold will be pruned).
  
 ===== Model types ===== ===== Model types =====
Line 45: Line 33:
  
 Individual machine learning models can be trained based on the structure of a set of compounds and all transformation rules triggered by them. These models can predict probabilities for all transformations of a new compound. As the learning problem clearly is a so called multi-label classification problem, we extended the machine learning approach using multi-label classifiers to improve the prediction. Individual machine learning models can be trained based on the structure of a set of compounds and all transformation rules triggered by them. These models can predict probabilities for all transformations of a new compound. As the learning problem clearly is a so called multi-label classification problem, we extended the machine learning approach using multi-label classifiers to improve the prediction.
- 
-Retrieved from "https://wiki.envipath.org/index.php?title=Relative_Reasoning&oldid=221" 
  
relative_reasoning.1710838107.txt.gz · Last modified: 2024/03/19 08:48 by shankar