Limits...
SpliceNet: recovering splicing isoform-specific differential gene networks from RNA-Seq data of normal and diseased samples.

Yalamanchili HK, Li Z, Wang P, Wong MP, Yao J, Wang J - Nucleic Acids Res. (2014)

Bottom Line: It eases the sample size bottleneck; evaluations on simulated data and lung cancer-specific ERBB2 and MAPK signaling pathways, with varying number of samples, evince the merit in handling high exon to sample size ratio datasets.Gene level evaluations demonstrate a substantial performance of SpliceNet over canonical correlation analysis, a method that is currently applied to exon level RNA-Seq data.SpliceNet can also be applied to exon array data.

View Article: PubMed Central - PubMed

Affiliation: Department of Biochemistry, The University of Hong Kong, Hong Kong (SAR), China Department of Pathology, The University of Hong Kong, Hong Kong (SAR), China.

Show MeSH

Related in: MedlinePlus

(a) Gene level dependencies from normal and cancer samples. (b) Differential edges of the same.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4150760&req=5

Figure 5: (a) Gene level dependencies from normal and cancer samples. (b) Differential edges of the same.

Mentions: The key merit of SpliceNet is in handling large dimensional data, where the number of exons per gene is large and comparable to sample size i.e. when the ratio of number of exons per gene to sample size is large. Firstly, to thoroughly evaluate the performance and stability of SpliceNet, simulations are performed by varying number of exons (dimensions) and samples. The performance of existing R package, RNASeqNet is also evaluated on the same data. The results summarized in Table 1 demonstrate the competence of SpliceNet in abstracting dependencies from exon-expression (high-dimensional) data. Secondly, SpliceNet and RNASeqNet are evaluated on cancer-specific ERBB2 and MAPK signaling pathways from KEGG database with different number of samples. The results summarized in Figure 3 evince the merit of SpliceNet over RNASeqNet in handling low sample datasets. Further, to appreciate the insights of differential cancer networks and their applications, a detailed work out of SpliceNet on Bcl-x and EGFR centered network is illustrated (Figures 4 and 5). Differential edges inferred by SpliceNet converged to cancer-specific splice variants reported in literature. Finally, to demonstrate the practical pertinence, performance of SpliceNet is also evaluated on real RNA-Seq data from three different tissues viz. lung, kidney and liver, alongside RNASeqNet. The F-scores reported in Table 4 demonstrate a significantly enhanced performance of SpliceNet over RNASeqNet.


SpliceNet: recovering splicing isoform-specific differential gene networks from RNA-Seq data of normal and diseased samples.

Yalamanchili HK, Li Z, Wang P, Wong MP, Yao J, Wang J - Nucleic Acids Res. (2014)

(a) Gene level dependencies from normal and cancer samples. (b) Differential edges of the same.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4150760&req=5

Figure 5: (a) Gene level dependencies from normal and cancer samples. (b) Differential edges of the same.
Mentions: The key merit of SpliceNet is in handling large dimensional data, where the number of exons per gene is large and comparable to sample size i.e. when the ratio of number of exons per gene to sample size is large. Firstly, to thoroughly evaluate the performance and stability of SpliceNet, simulations are performed by varying number of exons (dimensions) and samples. The performance of existing R package, RNASeqNet is also evaluated on the same data. The results summarized in Table 1 demonstrate the competence of SpliceNet in abstracting dependencies from exon-expression (high-dimensional) data. Secondly, SpliceNet and RNASeqNet are evaluated on cancer-specific ERBB2 and MAPK signaling pathways from KEGG database with different number of samples. The results summarized in Figure 3 evince the merit of SpliceNet over RNASeqNet in handling low sample datasets. Further, to appreciate the insights of differential cancer networks and their applications, a detailed work out of SpliceNet on Bcl-x and EGFR centered network is illustrated (Figures 4 and 5). Differential edges inferred by SpliceNet converged to cancer-specific splice variants reported in literature. Finally, to demonstrate the practical pertinence, performance of SpliceNet is also evaluated on real RNA-Seq data from three different tissues viz. lung, kidney and liver, alongside RNASeqNet. The F-scores reported in Table 4 demonstrate a significantly enhanced performance of SpliceNet over RNASeqNet.

Bottom Line: It eases the sample size bottleneck; evaluations on simulated data and lung cancer-specific ERBB2 and MAPK signaling pathways, with varying number of samples, evince the merit in handling high exon to sample size ratio datasets.Gene level evaluations demonstrate a substantial performance of SpliceNet over canonical correlation analysis, a method that is currently applied to exon level RNA-Seq data.SpliceNet can also be applied to exon array data.

View Article: PubMed Central - PubMed

Affiliation: Department of Biochemistry, The University of Hong Kong, Hong Kong (SAR), China Department of Pathology, The University of Hong Kong, Hong Kong (SAR), China.

Show MeSH
Related in: MedlinePlus