Limits...
Comparative analyses between retained introns and constitutively spliced introns in Arabidopsis thaliana using random forest and support vector machine.

Mao R, Raj Kumar PK, Guo C, Zhang Y, Liang C - PLoS ONE (2014)

Bottom Line: We demonstrated that our proposed feature extraction approach was more accurate in effectively classifying RIs from CSIs in comparison with other four approaches.Not only the basic sequence features and positional distribution characteristics of RIs were obtained, but also putative regulatory motifs in intron splicing were predicted based on our feature extraction approach.Clearly, our study will facilitate a better understanding of underlying mechanisms involved in intron retention.

View Article: PubMed Central - PubMed

Affiliation: College of Mechanical and Electronic Engineering, Northwest A&F University, Yangling, Shaanxi, China; College of Information Engineering, Northwest A&F University, Yangling, Shaanxi, China; Department of Biology, Miami University, Oxford, Ohio, United States of America.

ABSTRACT
One of the important modes of pre-mRNA post-transcriptional modification is alternative splicing. Alternative splicing allows creation of many distinct mature mRNA transcripts from a single gene by utilizing different splice sites. In plants like Arabidopsis thaliana, the most common type of alternative splicing is intron retention. Many studies in the past focus on positional distribution of retained introns (RIs) among different genic regions and their expression regulations, while little systematic classification of RIs from constitutively spliced introns (CSIs) has been conducted using machine learning approaches. We used random forest and support vector machine (SVM) with radial basis kernel function (RBF) to differentiate these two types of introns in Arabidopsis. By comparing coordinates of introns of all annotated mRNAs from TAIR10, we obtained our high-quality experimental data. To distinguish RIs from CSIs, We investigated the unique characteristics of RIs in comparison with CSIs and finally extracted 37 quantitative features: local and global nucleotide sequence features of introns, frequent motifs, the signal strength of splice sites, and the similarity between sequences of introns and their flanking regions. We demonstrated that our proposed feature extraction approach was more accurate in effectively classifying RIs from CSIs in comparison with other four approaches. The optimal penalty parameter C and the RBF kernel parameter [Formula: see text] in SVM were set based on particle swarm optimization algorithm (PSOSVM). Our classification performance showed F-Measure of 80.8% (random forest) and 77.4% (PSOSVM). Not only the basic sequence features and positional distribution characteristics of RIs were obtained, but also putative regulatory motifs in intron splicing were predicted based on our feature extraction approach. Clearly, our study will facilitate a better understanding of underlying mechanisms involved in intron retention.

Show MeSH
The mean occurrences of B frequent motifs between RIs and CSIs.In the left side of the histogram there are ten frequent motifs that have higher occurrences in RIs than in CSIs. In the right site of the histogram there are nine frequent motifs that have higher occurrences in CSIs than in RIs.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4128822&req=5

pone-0104049-g006: The mean occurrences of B frequent motifs between RIs and CSIs.In the left side of the histogram there are ten frequent motifs that have higher occurrences in RIs than in CSIs. In the right site of the histogram there are nine frequent motifs that have higher occurrences in CSIs than in RIs.

Mentions: Meanwhile, we also found that some short motifs were relatively frequent but quite different between the RIs and CSIs. So we extracted B feature set, and the results showed that they indeed helped us improve the classification performance, especially by using random forest (e.g., 2.3% F-Measure and Accuracy increase for our A+B+C feature set in comparison with A+C feature set, see Table 5). As showed in Figure 6, some short motifs (e.g., cc, gg, cg, ccg, cga, cgg, ggag, gggt, gaag, ttcg) have higher frequencies in RIs than CSIs whereas others (e.g., ta, at, atgt, taat, tatat, atatt, aaata, ttata, attat) occur higher frequencies in CSIs than RIs.


Comparative analyses between retained introns and constitutively spliced introns in Arabidopsis thaliana using random forest and support vector machine.

Mao R, Raj Kumar PK, Guo C, Zhang Y, Liang C - PLoS ONE (2014)

The mean occurrences of B frequent motifs between RIs and CSIs.In the left side of the histogram there are ten frequent motifs that have higher occurrences in RIs than in CSIs. In the right site of the histogram there are nine frequent motifs that have higher occurrences in CSIs than in RIs.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4128822&req=5

pone-0104049-g006: The mean occurrences of B frequent motifs between RIs and CSIs.In the left side of the histogram there are ten frequent motifs that have higher occurrences in RIs than in CSIs. In the right site of the histogram there are nine frequent motifs that have higher occurrences in CSIs than in RIs.
Mentions: Meanwhile, we also found that some short motifs were relatively frequent but quite different between the RIs and CSIs. So we extracted B feature set, and the results showed that they indeed helped us improve the classification performance, especially by using random forest (e.g., 2.3% F-Measure and Accuracy increase for our A+B+C feature set in comparison with A+C feature set, see Table 5). As showed in Figure 6, some short motifs (e.g., cc, gg, cg, ccg, cga, cgg, ggag, gggt, gaag, ttcg) have higher frequencies in RIs than CSIs whereas others (e.g., ta, at, atgt, taat, tatat, atatt, aaata, ttata, attat) occur higher frequencies in CSIs than RIs.

Bottom Line: We demonstrated that our proposed feature extraction approach was more accurate in effectively classifying RIs from CSIs in comparison with other four approaches.Not only the basic sequence features and positional distribution characteristics of RIs were obtained, but also putative regulatory motifs in intron splicing were predicted based on our feature extraction approach.Clearly, our study will facilitate a better understanding of underlying mechanisms involved in intron retention.

View Article: PubMed Central - PubMed

Affiliation: College of Mechanical and Electronic Engineering, Northwest A&F University, Yangling, Shaanxi, China; College of Information Engineering, Northwest A&F University, Yangling, Shaanxi, China; Department of Biology, Miami University, Oxford, Ohio, United States of America.

ABSTRACT
One of the important modes of pre-mRNA post-transcriptional modification is alternative splicing. Alternative splicing allows creation of many distinct mature mRNA transcripts from a single gene by utilizing different splice sites. In plants like Arabidopsis thaliana, the most common type of alternative splicing is intron retention. Many studies in the past focus on positional distribution of retained introns (RIs) among different genic regions and their expression regulations, while little systematic classification of RIs from constitutively spliced introns (CSIs) has been conducted using machine learning approaches. We used random forest and support vector machine (SVM) with radial basis kernel function (RBF) to differentiate these two types of introns in Arabidopsis. By comparing coordinates of introns of all annotated mRNAs from TAIR10, we obtained our high-quality experimental data. To distinguish RIs from CSIs, We investigated the unique characteristics of RIs in comparison with CSIs and finally extracted 37 quantitative features: local and global nucleotide sequence features of introns, frequent motifs, the signal strength of splice sites, and the similarity between sequences of introns and their flanking regions. We demonstrated that our proposed feature extraction approach was more accurate in effectively classifying RIs from CSIs in comparison with other four approaches. The optimal penalty parameter C and the RBF kernel parameter [Formula: see text] in SVM were set based on particle swarm optimization algorithm (PSOSVM). Our classification performance showed F-Measure of 80.8% (random forest) and 77.4% (PSOSVM). Not only the basic sequence features and positional distribution characteristics of RIs were obtained, but also putative regulatory motifs in intron splicing were predicted based on our feature extraction approach. Clearly, our study will facilitate a better understanding of underlying mechanisms involved in intron retention.

Show MeSH