Limits...
A mixture of feature experts approach for protein-protein interaction prediction.

Qi Y, Klein-Seetharaman J, Bar-Joseph Z - BMC Bioinformatics (2007)

Bottom Line: High-throughput methods can directly detect the set of interacting proteins in model species but the results are often incomplete and exhibit high false positive and false negative rates.However, due to missing data and high redundancy among the features used, different protein pairs may benefit from different features based on the set of attributes available.Our method improved upon the best previous methods for this task.

View Article: PubMed Central - HTML - PubMed

Affiliation: School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213, USA. qyj@cs.cmu.edu

ABSTRACT

Background: High-throughput methods can directly detect the set of interacting proteins in model species but the results are often incomplete and exhibit high false positive and false negative rates. A number of researchers have recently presented methods for integrating direct and indirect data for predicting interactions. These methods utilize a common classifier for all pairs. However, due to missing data and high redundancy among the features used, different protein pairs may benefit from different features based on the set of attributes available. In addition, in many cases it is hard to directly determine which of the data sources contributed to a prediction. This information is important for biologists using these predications in the design of new experiments.

Results: To address these challenges we propose a Mixture-of-Feature-Experts method for protein-protein interaction prediction. We split the features into roughly homogeneous sets of feature experts. The individual experts use logistic regression and their scores are combined using another logistic regression. When combining the scores the weighting of each expert depends on the set of input attributes available for that pair. Thus, different experts will have different influence on the prediction depending on the available features.

Conclusion: We applied our method to predict the set of interacting proteins in yeast and human cells. Our method improved upon the best previous methods for this task. In addition, the weighting of the experts provides means to evaluate the prediction based on the high scoring features.

Show MeSH

Related in: MedlinePlus

Mixture of Four Feature Experts in Yeast. Graphical representation of the Mixture-of-Feature-Experts method (MFE) for yeast. Table 1 lists the features used by each of the four experts. For definition of P, F, S, E experts, see details in the 'Feature' section.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2230507&req=5

Figure 2: Mixture of Four Feature Experts in Yeast. Graphical representation of the Mixture-of-Feature-Experts method (MFE) for yeast. Table 1 lists the features used by each of the four experts. For definition of P, F, S, E experts, see details in the 'Feature' section.

Mentions: We design a method called Mixture-of-Feature-Experts (MFE) to achieve the above computational properties. As Figure 2 shows, our framework can be viewed as a single layer tree, with feature experts at the leaves. Each expert uses one of the dataset groups to predict PPIs. A root gate is used to integrate predictions from multiple feature experts. The weights assigned to each of the experts by the root gate depends on the input set for a given pair. Intuitively, this framework is analogous to the following process: each feature expert gives their opinion about how likely the investigated pair interacts and then the gate creates a final decision by the weighted sum of the experts' predictions. Moreover, these weights are local and specific to the current example pair.


A mixture of feature experts approach for protein-protein interaction prediction.

Qi Y, Klein-Seetharaman J, Bar-Joseph Z - BMC Bioinformatics (2007)

Mixture of Four Feature Experts in Yeast. Graphical representation of the Mixture-of-Feature-Experts method (MFE) for yeast. Table 1 lists the features used by each of the four experts. For definition of P, F, S, E experts, see details in the 'Feature' section.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2230507&req=5

Figure 2: Mixture of Four Feature Experts in Yeast. Graphical representation of the Mixture-of-Feature-Experts method (MFE) for yeast. Table 1 lists the features used by each of the four experts. For definition of P, F, S, E experts, see details in the 'Feature' section.
Mentions: We design a method called Mixture-of-Feature-Experts (MFE) to achieve the above computational properties. As Figure 2 shows, our framework can be viewed as a single layer tree, with feature experts at the leaves. Each expert uses one of the dataset groups to predict PPIs. A root gate is used to integrate predictions from multiple feature experts. The weights assigned to each of the experts by the root gate depends on the input set for a given pair. Intuitively, this framework is analogous to the following process: each feature expert gives their opinion about how likely the investigated pair interacts and then the gate creates a final decision by the weighted sum of the experts' predictions. Moreover, these weights are local and specific to the current example pair.

Bottom Line: High-throughput methods can directly detect the set of interacting proteins in model species but the results are often incomplete and exhibit high false positive and false negative rates.However, due to missing data and high redundancy among the features used, different protein pairs may benefit from different features based on the set of attributes available.Our method improved upon the best previous methods for this task.

View Article: PubMed Central - HTML - PubMed

Affiliation: School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213, USA. qyj@cs.cmu.edu

ABSTRACT

Background: High-throughput methods can directly detect the set of interacting proteins in model species but the results are often incomplete and exhibit high false positive and false negative rates. A number of researchers have recently presented methods for integrating direct and indirect data for predicting interactions. These methods utilize a common classifier for all pairs. However, due to missing data and high redundancy among the features used, different protein pairs may benefit from different features based on the set of attributes available. In addition, in many cases it is hard to directly determine which of the data sources contributed to a prediction. This information is important for biologists using these predications in the design of new experiments.

Results: To address these challenges we propose a Mixture-of-Feature-Experts method for protein-protein interaction prediction. We split the features into roughly homogeneous sets of feature experts. The individual experts use logistic regression and their scores are combined using another logistic regression. When combining the scores the weighting of each expert depends on the set of input attributes available for that pair. Thus, different experts will have different influence on the prediction depending on the available features.

Conclusion: We applied our method to predict the set of interacting proteins in yeast and human cells. Our method improved upon the best previous methods for this task. In addition, the weighting of the experts provides means to evaluate the prediction based on the high scoring features.

Show MeSH
Related in: MedlinePlus