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

Graphical Model View of Mixture-of-Experts (ME) Method. A graphical model view of the Mixture-of-Experts (ME) method. The target variable Y is dependent on the input vector X and the multinomial random variable M. P(M/X) is modeled by the gate while P(Y/X, M) is modeled by the experts.
© Copyright Policy - open-access
Related In: Results  -  Collection

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

Figure 3: Graphical Model View of Mixture-of-Experts (ME) Method. A graphical model view of the Mixture-of-Experts (ME) method. The target variable Y is dependent on the input vector X and the multinomial random variable M. P(M/X) is modeled by the gate while P(Y/X, M) is modeled by the experts.

Mentions: where M is a set of hidden data and indicates which expert was responsible for generating each example data pair. Having I experts, M is a I-dimensional indicator vector variable. That is, all entries in M are 0 except for one of the entries which is set to 1. The sum is over all configuration of variable M. In other words, target class label Y is dependent on the input data X and the choice of expert M. The choice of M is also dependent on the input X. P(M/X) is modeled using the root gate, while P(Y/M, X) is modeled by each feature expert in our framework. The graphical model view of MFE method is illustrated in Figure 3. This Bayesian network structure expresses that the target variable Y is dependent on the input vector variable X and the multinomial random variable M. It is essentially a modification of the probabilistic Mixture-of-Experts (ME) model [32].


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

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

Graphical Model View of Mixture-of-Experts (ME) Method. A graphical model view of the Mixture-of-Experts (ME) method. The target variable Y is dependent on the input vector X and the multinomial random variable M. P(M/X) is modeled by the gate while P(Y/X, M) is modeled by the experts.
© Copyright Policy - open-access
Related In: Results  -  Collection

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

Figure 3: Graphical Model View of Mixture-of-Experts (ME) Method. A graphical model view of the Mixture-of-Experts (ME) method. The target variable Y is dependent on the input vector X and the multinomial random variable M. P(M/X) is modeled by the gate while P(Y/X, M) is modeled by the experts.
Mentions: where M is a set of hidden data and indicates which expert was responsible for generating each example data pair. Having I experts, M is a I-dimensional indicator vector variable. That is, all entries in M are 0 except for one of the entries which is set to 1. The sum is over all configuration of variable M. In other words, target class label Y is dependent on the input data X and the choice of expert M. The choice of M is also dependent on the input X. P(M/X) is modeled using the root gate, while P(Y/M, X) is modeled by each feature expert in our framework. The graphical model view of MFE method is illustrated in Figure 3. This Bayesian network structure expresses that the target variable Y is dependent on the input vector variable X and the multinomial random variable M. It is essentially a modification of the probabilistic Mixture-of-Experts (ME) model [32].

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