Intron identification approaches based on weighted features and fuzzy decision trees

Huang, Yin-Fu; Liang, Ching-Ping; Liou, Sing-Wu

doi:10.1016/j.compbiomed.2011.10.015

Cited by 4 publications

(2 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some researchers (Chen, Liu, Vanschoenwinkel, & Manderick, 2009) proposed a context-based approach statistical method incorporated with SVM for accurate prediction of human genome motif sequence. Huang, Liang, and Liou (2012) introduced a hybrid approach based on fuzzy decision trees. Fuzzy rules were generated using unsupervised self-organizing map along with gain ratio estimation.…”

Section: Introductionmentioning

confidence: 99%

A novel approach for accurate identification of splice junctions based on hybrid algorithms

Mandal

2014

Journal of Biomolecular Structure and Dynamics

View full text Add to dashboard Cite

The precise prediction of splice junctions as 'exon-intron' or 'intron-exon' boundaries in a given DNA sequence is an important task in Bioinformatics. The main challenge is to determine the splice sites in the coding region. Due to the intrinsic complexity and the uncertainty in gene sequence, the adoption of data mining methods is increasingly becoming popular. There are various methods developed on different strategies in this direction. This article focuses on the construction of new hybrid machine learning ensembles that solve the splice junction task more effectively. A novel supervised feature reduction technique is developed using entropy-based fuzzy rough set theory optimized by greedy hill-climbing algorithm. The average prediction accuracy achieved is above 98% with 95% confidence interval. The performance of the proposed methods is evaluated using various metrics to establish the statistical significance of the results. The experiments are conducted using various schemes with human DNA sequence data. The obtained results are highly promising as compared with the state-of-the-art approaches in literature.

show abstract

Section: Introductionmentioning

confidence: 99%

A novel approach for accurate identification of splice junctions based on hybrid algorithms

Mandal

2014

Journal of Biomolecular Structure and Dynamics

View full text Add to dashboard Cite

show abstract

“…The extracted features are usually based on nucleotide position information [4–9], the frequency of k -mers [4, 6, 10], dependence between adjacent and nonadjacent nucleotides [1, 6, 11–13], RNA secondary structure information [14–18], DNA structural properties [19], and some other attributes that can be calculated directly from sequence information [20–22]. The commonly used classifiers include support vector machine (SVM) [1, 3, 5, 6, 10, 18, 23–25], artificial neural network (ANN) [26–29], random forest (RF) [13], and decision tree [30].…”

Section: Introductionmentioning

confidence: 99%

A high-performance approach for predicting donor splice sites based on short window size and imbalanced large samples

Zeng

Yuan

et al. 2019

Biol Direct

View full text Add to dashboard Cite

Background Splice sites prediction has been a long-standing problem in bioinformatics. Although many computational approaches developed for splice site prediction have achieved satisfactory accuracy, further improvement in predictive accuracy is significant, for it is contributing to predict gene structure more accurately. Determining a proper window size before prediction is necessary. Overly long window size may introduce some irrelevant features, which would reduce predictive accuracy, while the use of short window size with maximum information may performs better in terms of predictive accuracy and time cost. Furthermore, the number of false splice sites following the GT–AG rule far exceeds that of true splice sites, accurate and rapid prediction of splice sites using imbalanced large samples has always been a challenge. Therefore, based on the short window size and imbalanced large samples, we developed a new computational method named chi-square decision table (χ 2 -DT) for donor splice site prediction. Results Using a short window size of 11 bp, χ 2 -DT extracts the improved positional features and compositional features based on chi-square test, then introduces features one by one based on information gain, and constructs a balanced decision table aimed at implementing imbalanced pattern classification. With a 2000:271,132 (true sites:false sites) training set, χ 2 -DT achieves the highest independent test accuracy (93.34%) when compared with three classifiers (random forest, artificial neural network, and relaxed variable kernel density estimator) and takes a short computation time (89 s). χ 2 -DT also exhibits good independent test accuracy (92.40%), when validated with BG-570 mutated sequences with frameshift errors (nucleotide insertions and deletions). Moreover, χ 2 -DT is compared with the long-window size-based methods and the short-window size-based methods, and is found to perform better than all of them in terms of predictive accuracy. Conclusions Based on short window size and imbalanced large samples, the proposed method not only achieves higher predictive accuracy than some existing methods, but also has high computational speed and good robustness against nucleotide insertions and deletions. Reviewers This article was reviewed by Ryan McGinty, Ph.D. and Dirk Walther. Electronic supplementary material The online version of this article (10.1186/s13062-019-0236-y) contains supplementary material, which is available to authorized users.

show abstract

A novel approach for predicting DNA splice junctions using hybrid machine learning algorithms

Mandal

2014

Soft Comput

View full text Add to dashboard Cite

Intron identification approaches based on weighted features and fuzzy decision trees

Cited by 4 publications

References 24 publications

A novel approach for accurate identification of splice junctions based on hybrid algorithms

A novel approach for accurate identification of splice junctions based on hybrid algorithms

A high-performance approach for predicting donor splice sites based on short window size and imbalanced large samples

A novel approach for predicting DNA splice junctions using hybrid machine learning algorithms

Contact Info

Product

Resources

About