Identification of rare alternative splicing events in MS/MS data reveals a significant fraction of alternative translation initiation sites

PeerJ. 2014 Nov 13:2:e673. doi: 10.7717/peerj.673. eCollection 2014.

Abstract

Integration of transcriptome data is a crucial step for the identification of rare protein variants in mass-spectrometry (MS) data with important consequences for all branches of biotechnology research. Here, we used Splooce, a database of splicing variants recently developed by us, to search MS data derived from a variety of human tumor cell lines. More than 800 new protein variants were identified whose corresponding MS spectra were specific to protein entries from Splooce. Although the types of splicing variants (exon skipping, alternative splice sites and intron retention) were found at the same frequency as in the transcriptome, we observed a large variety of modifications at the protein level induced by alternative splicing events. Surprisingly, we found that 40% of all protein modifications induced by alternative splicing led to the use of alternative translation initiation sites. Other modifications include frameshifts in the open reading frame and inclusion or deletion of peptide sequences. To make the dataset generated here available to the community in a more effective form, the Splooce portal (http://www.bioinformatics-brazil.org/splooce) was modified to report the alternative splicing events supported by MS data.

Keywords: Alternative splicing events; Mass spectrometry; Peptide identification; Proteomics; Translation initiation sites.

Grants and funding

JEK is supported by a post-doctoral fellowship from CNPq (501891/2013-7). This research was supported by grants from CNPq (483775/2012-6) and CAPES (edital 051/2013), both to SJS. GAdS is a Special Visiting Scientist to the Brain Institute – UFRN (supported by a CNPq grant 400392/2014-3 to SJS). GAdS and the Proteomics Core Facility are supported by grants from UiO and the Norwegian South-East Health Authority (Helse Sør-Øst). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.