Chapter 25 Genome Annotation

I don’t intend this to be a treatise on how to actually annotate a genome. Presumably, that is a task that involves feeding a genome and a lot of mRNA transcripts into a pipeline that then makes gene models, etc. I guess I could talk a little about that process, ’cuz it would be fun to learn more about it.

However, I will be more interested in understanding what annotation data look like (i.e. in a GFF file) and how to associate it with SNP data (i.e. using snpEff).

The GFF format is a distinctly hierarchical format, but it is still tabular, it is not in XML, thank god! ’cuz it is much easier to parse in tabular format.

You can fiddle it with bedtools.

Here is an idea for a fun thing for me to do: Take a big chunk of chinook GFF (and maybe a few other species), and then figure out who the parents are of each of the rows, and then make a graph (with dot) showing all the different links (i.e. gene -> mRNA -> exon -> CDS) etc, and count up the number of occurrences of each, in order to get a sense of what sorts of hierarchies a typical GFF file contains.