Abstract:Motivation: The network of interactions between transcription factors and their regulatory gene targets governs many of the behaviors and responses of cells. The construction of regulatory network have been decomposed to identify, for every known regulator, its target genes, its binding motif and its DNA binding sites. Many tools have been developed in the last decade to solve these problems. However, Tompa et al. (2005) showed that the performance of individual algorithm was not constantly good for all transcription factors. Because machine learning algorithms have shown advantage in integrating information of different types, we exploit this property in the integration of predictions from an ensemble of commonly used motif exploration algorithms. Results: We selected In this paper, we introduced 3 ensemble machine methods to integrated the predictions from 5 commonly used motif exploration algorithms. Besides the conventional PWM scanning model, we developed w-scanning model, which uses the feature importance measure abstracted from a binary classification as model to identify significant (overrepresented) k-mers and potential binding site across the entire genome. With both rescanning models, the comprehensive ensemble machine provided an better alternative to the conventional PWM model in the DNA binding analysis. We also introduced the definition of PWM k-mer subspaces, which provide a dimension reduction tool for sequence analysis and, to some degree, enables machine learning method to success in small sample situation. We test the performance in identifying gene targets and binding motifs over 88 yeast transcription factors. The ensemble method is able to integrate the orthogonal information from different weak learners to method and to perform consistently well for more transcription factors. It is useful in completing the transcription regulatory network for the entire genome. Note the ensemble is easily extended to include more tools as well as more information in the

A Combined Model and a Varied Gibbs Sampling Algorithm Used for Motif Discovery.

An Improved Gibbs Sampling Method for Motif Finding

An Improved Gibbs Sampling Method for Motif Discovery Via Sequence Weighting.

GSMC: Combining Parallel Gibbs Sampling with Maximal Cliques for Hunting DNA Motif

A Greedy Two-stage Gibbs Sampling Method for Motif Discovery in Biological Sequences

An Improved Genetic Algorithm For Dna Motif Discovery With Public Domain Information

W-AlignACE: an Improved Gibbs Sampling Algorithm Based on More Accurate Position Weight Matrices Learned from Sequence and Gene Expression/chip-Chip Data.

A Two-Block Motif Discovery Method With Improved Accuracy

An Improved Algorithm On Detecting Transcription And Translation Motif In Archaeal Genomic Sequences

A Comprehensive Survey on Genetic Algorithms for DNA Motif Prediction

Multiple Motif Discovery in Biological Sequences by Mixture Gibbs Sampling

MotifHub: Detection of trans-acting DNA motif group with probabilistic modeling algorithm

An Algorithm for Motif Discovery with Iteration on Lengths of Motifs

A higher-order background model improves the detection of promoter regulatory elements by Gibbs sampling

Computational Discovery of Regulatory DNA Motifs Using Evolutionary Computation

Ensemble Machine Methods for Analysis of Transcription Factor and DNA Interactions

Igapk: Improved Gapk Algorithm For Regulatory Dna Motif Discovery

Discovering DNA shape motifs with multiple DNA shape features: generalization, methods, and validation

Position Dependencies Among Multinomial Distribution For Motif Discovery Based On Gibbs Sampling

Multi-motif Identification Using Differential Information Content and Cluster Refine

Coupling hidden Markov models for the discovery of Cis-regulatory modules in multiple species