Premium
Identification of novel prognostic alternative splicing signature in papillary renal cell carcinoma
Author(s) -
Duan Yi,
Zhang Dong
Publication year - 2020
Publication title -
journal of cellular biochemistry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.028
H-Index - 165
eISSN - 1097-4644
pISSN - 0730-2312
DOI - 10.1002/jcb.29314
Subject(s) - papillary renal cell carcinomas , rna splicing , proportional hazards model , alternative splicing , oncology , gene expression profiling , medicine , univariate , computational biology , biology , bioinformatics , cancer research , multivariate statistics , renal cell carcinoma , gene , gene expression , genetics , computer science , messenger rna , rna , machine learning
Papillary renal cell carcinoma (pRCC) is a heterogeneous disease containing multifocal or solitary tumors with an aggressive phenotype. Increasing evidence has indicated the involvement of aberrant splicing variants in renal cell cancer, while systematic profiling of aberrant alternative splicing (AS) in pRCC was lacking and largely unknown. In the current study, comprehensive profiling of AS events were performed based on the integration of pRCC cohort from the Cancer Genome Atlas database and SpliceSeq software. With rigorous screening and univariate Cox analysis, a total of 2077 prognoses AS events from 1642 parent genes were identified. Then, stepwise least absolute shrinkage and selection operator method‐penalized Cox regression analyses with 10‐fold cross‐validation followed by multivariate Cox regression were used to construct the prognostic AS signatures within each AS type. And a final 21 AS event‐based signature was proposed which showed potent prognostic capability in stratifying patients into low‐ and high‐risk subgroups ( P < .0001). Furthermore, time‐dependent receiver operating characteristics curves confirmed that the final AS signature was effective and robust in predicting overall survival for pRCC patients with the area under the curve above 0.9 from 1 to 5 years. In addition, splicing correlation network was built to uncover the potential regulatory pattern among prognostic splicing factors and candidate AS events. Besides, gene set enrichment analysis revealed the involvement of these candidates AS events in tumor‐related pathways including extracellular matrix organization, oxidative phosphorylation, and P53 signaling pathways. Taken together, our results could contribute to elucidating the underlying mechanism of AS in the oncogenesis process and broaden the novel field of prognostic and clinical application of molecule‐targeted approaches in pRCC.