Revealing differences in gene network inference algorithms on the network level by ensemble methods

Altay, Gökmen; Emmert‐Streib, Frank

doi:10.1093/bioinformatics/btq259

Cited by 93 publications

(96 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Assessments of methods are performed using metrics, usually with area under receiver operating characteristics (AUROC), F-score, or area under precision-recall curve (AUPR) (Altay and Emmert-Streib, 2010b;Narendra et al, 2011). Nonetheless, in our approach, since the interaction databases of the literature are far from being complete, these may cause too many FPs and thus are not suitable metrics.…”

Section: Resultsmentioning

confidence: 99%

Global assessment of network inference algorithms based on available literature of gene/protein interactions

Altay¹,

N²,

Neal³

2013

Turk J Biol

Self Cite

View full text Add to dashboard Cite

We propose a framework that uses the available gene/protein interaction databases of the literature as a universal benchmark in order to globally assess the inference performances of gene network inference algorithms. We also developed an R software package for convenient use of the framework, which can also be used in general as a quick tool to search in the literature for available validations of interactions. We applied the proposed approach to 2 publicly available prostate cancer gene expression datasets and a large breast cancer gene expression dataset. The results revealed different aspects and superiority of algorithms that had not been compared previously in the available literature. Our approach allowed the assessing and comparing of the algorithms on a real dataset of a size of around 30,000 probes, which showed the strengths and weaknesses of the algorithms from different points of view rather than conventional approaches. We further show that our approach provides a unique advantage in assessing the performance of an inference method when applied to a new dataset and thus sheds light on the results of a de novo application, which would be obscure without our approach.

show abstract

Section: Resultsmentioning

confidence: 99%

Global assessment of network inference algorithms based on available literature of gene/protein interactions

Altay¹,

N²,

Neal³

2013

Turk J Biol

Self Cite

View full text Add to dashboard Cite

show abstract

“…[12,[69][70][71]. In order to detect truly coexpressed gene pairs in an ad-hoc way, the calculated correlation values are compared with a predefined correlation cut-off value.…”

Section: Relevance Networkmentioning

confidence: 99%

A Survey of Computational Approaches to Reconstruct and Partition Biological Networks

Acharya

Judeh

Zhu

2012

Statistical and Machine Learning Approaches for Network Analysis

View full text Add to dashboard Cite

“…So the methods inevitably make assumptions when dealing with gene expression datasets, which cause different types of systematic inference bias (De Smet & Marchal, 2010). For example, mutual information based algorithms such as ARANCE (Margolin et al, 2006), CLR (Faith et al, 2007), MRNET (Meyer et al, 2007) and Relevance Networks (Butte & Kohane, 2000) systematically discriminate activating interactions and bias towards repressing interactions (Altay & Emmert-Streib, 2010).…”

Section: Introductionmentioning

confidence: 99%

A mixture-of-experts approach for gene regulatory network inference

Shao

Lavesson

Boeva

et al. 2016

IJDMB

View full text Add to dashboard Cite

Context. Gene regulatory network (GRN) inference is an important and challenging problem in bioinformatics. A variety of machine learning algorithms have been applied to increase the GRN inference accuracy. Ensemble learning methods are shown to yield a higher inference accuracy than individual algorithms. Objectives. We propose an ensemble GRN inference method, which is based on the principle of Mixture-of-Experts ensemble learning. The proposed method can quantitatively measure the accuracy of individual GRN inference algorithms at the network motifs level. Based on the accuracy of the individual algorithms at predicting different types of network motifs, weights are assigned to the individual algorithms so as to take advantages of their strengths and weaknesses. In this way, we can improve the accuracy of the ensemble prediction. Methods. The research methodology is controlled experiment. The independent variable is method. It has eight groups: five individual algorithms, the generic average ranking method used in the DREAM5 challenge, the proposed ensemble method including four types of network motifs and five types of network motifs. The dependent variable is GRN inference accuracy, measured by the area under the precision-recall curve (AUPR). The experiment has training and testing phases. In the training phase, we analyze the accuracy of five individual algorithms at the network motifs level to decide their weights. In the testing phase, the weights are used to combine predictions from the five individual algorithms to generate ensemble predictions. We compare the accuracy of the eight method groups on Escherichia coli microarray dataset using AUPR. Results. In the training phase, we obtain the AUPR values of the five individual algorithms at predicting each type of the network motifs. In the testing phase, we collect the AUPR values of the eight methods on predicting the GRN of the Escherichia coli microarray dataset. Each method group has a sample size of ten (ten AUPR values). Conclusions. Statistical tests on the experiment results show that the proposed method yields a significantly higher accuracy than the generic average ranking method. In addition, a new type of network motif is found in GRN, the inclusion of which can increase the accuracy of the proposed method significantly.

show abstract

Revealing differences in gene network inference algorithms on the network level by ensemble methods

Cited by 93 publications

References 25 publications

Global assessment of network inference algorithms based on available literature of gene/protein interactions

Global assessment of network inference algorithms based on available literature of gene/protein interactions

A Survey of Computational Approaches to Reconstruct and Partition Biological Networks

A mixture-of-experts approach for gene regulatory network inference

Contact Info

Product

Resources

About