Annotation of Alternatively Spliced Proteins and Transcripts with Protein-Folding Algorithms and Isoform-Level Functional Networks

Abstract

Tens of thousands of splice isoforms of proteins have been catalogued as predicted sequences from transcripts in humans and other species. Relatively few have been characterized biochemically or structurally. With the extensive development of protein bioinformatics, the characterization and modeling of isoform features, isoform functions, and isoform-level networks have advanced notably. Here we present applications of the I-TASSER family of algorithms for folding and functional predictions and the IsoFunc, MIsoMine, and Hisonet data resources for isoform-level analyses of network and pathway-based functional predictions and protein-protein interactions. Hopefully, predictions and insights from protein bioinformatics will stimulate many experimental validation studies.

Keywords: Functional prediction; Isoform network; Protein folding; Splice isoforms.

Fig. 1

Flowcharts of template-based structure prediction by I-TASSER (a) and ab initio structural folding by QUARK (b)

Fig. 2

Protocol of COACH for structure-based protein function annotation. Prediction results from five different programs are combined using support vector machine to increase accuracy and coverage of function annotations

Fig. 3

The I-TASSER models for two splice isoforms of Annexin 6. (a) The “TPS” residues in anxa6-001 are exposed to solvent which helps increase the likelihood of phosphorylation as a post-translational modification. The hydroxyl groups, which are the target of kinases for phosphorylation, are highlighted in the inset. (b) Due to the absence of “VAAEIL” residues (aa 525–530 in anxa6-001) in the anxa6-002 variant, the “TPS” residues are either partially or completely buried by other atoms which significantly reduce the possibility of the protein for phosphorylation [Modified from reference [1]]

Fig. 4

The schematic of iterative multiple-instance learning (MIL) for isoform-level function prediction

Fig. 5

An example query of Anxa6 gene from the mouse isoform function database

Fig. 6

The functional networks of the highest connected isoform (HCI) (NM_002493.4, NDUFB6_2) and non- highest connected isoform (NCI) (NM_182739.2, NDUFB6_1) of the NADH dehydrogenase (ubiquinone) 1 beta subcomplex, 6 (NDUFB6) gene

Fig. 7

Isoform networks of ERBB2_1 (NM_004448.2) and ERBB2_2 (NM_001005862) of the human ERBB2 gene