[maker-devel] Few alternative isoforms when alt_splice=0

[Ju Roscito]

Dear all,

I have a question about the behaviour of alt_splice option, seems there’s not much about it on the forum.

I have run a single round of MAKER (2.31.9) on a vertebrate genome, with trinity mRNA data and mapped proteins from closely-related species. I set alt_splice to 0, but still got from two to four mRNAs for ~20 out of the 19.000 predicted genes. Has someone also seen the same? Any idea why would that happen?

Thanks a lot in advance.
_______________________________________________
maker-devel mailing list
maker...@box290.bluehost.com
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org

[Carson Holt]

The alt_splice=1 option will cause MAKER to generate separate sets of hints based on non-concordant transcript alignment evidence, and give them to the gene predictors (i.e. it run each predictor multiple times on the same locus, but with different hint sets). So you can sometimes get alternate models that way if transcript evidence is deep. When it’s off, each gene predictor gets a single set of hints (non-concordant hints are collapsed).

However Augustus sometimes decides to call multiple splicing all by itself (it’s internal to their algorithm). In that case you can still get a gene with multiple transcripts (if I remember right MAKER3 tries to suppress this behavior from Augustus, but MAKER2 does not).

—Carson