ISOGO: Functional annotation of protein-coding splice variants

Abstract

the advent of RnA-seq technologies has switched the paradigm of genetic analysis from a genome to a transcriptome-based perspective. Alternative splicing generates functional diversity in genes, but the precise functions of many individual isoforms are yet to be elucidated. Gene Ontology was developed to annotate gene products according to their biological processes, molecular functions and cellular components. Despite a single gene may have several gene products, most annotations are not isoform-specific and do not distinguish the functions of the different proteins originated from a single gene. Several approaches have tried to automatically annotate ontologies at the isoform level, but this has shown to be a daunting task. We have developed ISOGO (ISOform + GO function imputation), a novel algorithm to predict the function of coding isoforms based on their protein domains and their correlation of expression along 11,373 cancer patients. Combining these two sources of information outperforms previous approaches: it provides an area under precision-recall curve (AUPRC) five times larger than previous attempts and the median AUROC of assigned functions to genes is 0.82. We tested ISOGO predictions on some genes with isoform-specific functions (BRCA1, MADD,VAMP7 and ITSN1) and they were coherent with the literature. Besides, we examined whether the main isoform of each gene-as predicted by APPRIS-was the most likely to have the annotated gene functions and it occurs in 99.4% of the genes. We also evaluated the predictions for isoform-specific functions provided by the CAFA3 challenge and results were also convincing. To make these results available to the scientific community, we have deployed a web application to consult ISOGO predictions (https://biotecnun.unav. es/app/isogo). Initial data, website link, isoform-specific GO function predictions and R code is available at https://gitlab.com/icassol/isogo. Alternative splicing (AS) is a genetic process by which a single pre-mRNA can originate different mature mRNAs (called isoforms or transcripts) by including or excluding exons and introns 1-4. It is estimated that genes have on average 7 transcripts, that the whole transcriptome there are more than 100,000 AS events 5,6 and that over 90% of human genes contain one or more isoforms 7-10. From a functional point of view, AS is an intriguing process. Some studies show that a large number of sporadic splicing events produce alternative isoforms lowly expressed, and thus may be non-functional noise in the transcription process 11-13. On the other hand, other studies show and experimentally validate that different isoforms originated by alternative splicing may have distinct or even opposite functions 14,15. It is known that AS can cause cellular abnormalities that lead to diverse genetic diseases. All the hallmarks of cancer have their counterpart in AS 16-18. For example, BRCA1 is a tumor suppressor gene related to breast cancer susceptibility. Its isoform originated from skipping exon 11 (that includes a RAD51 inte...

Deciphering the Interregional and Interhemisphere Proteome of the Human Brain in the Context of the Human Proteome Project

“…UPEFinder () along with NeXtProt, PHAROS, and literature curation were mainly used for the uPE1 protein search. , The structure and function prediction workflow of uPE1 were taken from HPSF (Human Proteome Structure and Function) and the associated literature. , In brief, the FASTA sequence of the particular uPE1 protein was downloaded, and a protein BLAST () was performed to screen homologous protein. The protein was also searched in ISOGO for extracting isoform information based on the prediction . Furthermore, the FASTA sequence was uploaded in i-TASSER followed by COFACTOR for structure and function prediction. , Lastly, the structure and function were selected from the output based on the TM score, RMSD for i-TASSER, and Cscore GO for COFACTOR.…”

“…The protein was also searched in ISOGO for extracting isoform information based on the prediction. 31 Furthermore, the FASTA sequence was uploaded in i-TASSER followed by COFACTOR for structure and function prediction. 32,33 Lastly, the structure and function were selected from the output based on the TM score, RMSD for i-TASSER, and Cscore GO for COFACTOR.…”

FINER: enhancing the prediction of tissue-specific functions of isoforms by refining isoform interaction networks

“…We further make comprehensive comparisons between the functional prediction performance of FINER and that of several state-of-the-art methods with different objectives, including two recent isoform function prediction methods DIFFUSE ( 12 ) and DisoFun ( 14 ), a tissue-specific protein function prediction method OhmNet ( 20 ) and a general biological network refinement method NE ( 59 ). Note that three isoform function prediction methods, DisoFun, ISOGO ( 16 ) and IsoResolve ( 15 ) have been published in the literature after DIFFUSE. Although these methods have not been compared with DIFFUSE directly on the same dataset, their reported overall performance all seem to be worse than that of DIFFUSE.…”

Expression and Splicing Mediate Distinct Biological Signals

“…Since we find splicing is both frequent and significant, our tool provides an essential way to extend the functionality of these databases. Interestingly some people have started to work on transcript-level databases [44][45][46] , but there is still much to do.…”