Coding/Noncoding and Synonymous/Nonsynonymous

[P-Richmond]

Has anyone calculated, for a given genome (S288c), what the % of the genome is coding vs noncoding? Also, has anyone calculated for the coding regions, the % of positions that when mutated cause a synonymous vs nonsynonymous SNP?

If you can point me the right direction that would help.

Cheers,
Phil

[DZhang]

Phil,

For the first few question, you need a simple script to do the calculation based on the gene annotation. Or for a specific genome, all the ORF sequences may already exist in fasta, which makes the task easier. As to your last question, you need to think it more carefully.

Best regards,
Douglas

https://www.contigexpress.com

[TiborNagy]

I think it is very hard to calculate nonsynonymous SNPs, when some of the ORFs are overlap (source). Also very hard to calculate the coding vs noncoding percentages, because some of the ORFs are not verified (source).

[P-Richmond]

I've written a script that takes a reference fasta and a corresponding annotation gff3 and calculated for verified ORFs (non-overlapping) the synonymous vs. nonsynonymous percentages. For S288c, 6380 genes that fit into classical ORF descriptions (start, with stop in frame)--no spliced exons yet (sill working on that) the % of positions that have nonsynonymous potential is 74.9651%.

This script should be generalizable and I can give it to those interested when its completed.

-Phil