Exon-Spliced Graph vs Gene Model

Question

[isu2017]

Hi everyone,

I was wondering if anybody could clarify for this for me. I know of many papers where they create exon-spliced graphs based upon the output of software (such as Augustus.) My understanding was an exon-spliced graph just showcases the main isoform of a gene, in addition to any alternative splicing that could occur. Is that correct? Additionally, how does this differ from a gene model? Or do gene models generally only show the predominate isoform.

Answer 1

[pltbiotech_tkarthi]

I think, unless you give input of alternative splicing derived cDNA/mRNA sequences, the online software Augustus would predict a single isoform, starts from the beginning of the sequence wherever it would find the first CDS (ATG to TGA encodes for a protein) as a result. For predicting an alternatively spliced transcripts, you should have the experimentally derived mRNA sequences (alternatively spliced mRNA transcript sequences) from closely relevant species to predict bioinformatic based alternatively spliced sites from another related species. At least you should have the transcripts information within tribe of one species to predict the splice sites in another species. For instance, T.urartu to A. speltoides, they are in the same tribe: Triticeae.

Alternatively you could also try: Splign server from NCBI: https://www.ncbi.nlm.nih.gov/sutils/splign/splign.cgi?textpage=online&level=form But, here also you need to have cDNA/mRNA transcripts information to align with genomic DNA. For alternatively spliced transcripts, you have to give each transcript as separate input file to align with the genomic fragment of interest. Hope it may help.