We're working with the nomenclature found at Varnomen

The question is; is there a fully reliable algorithm for determining the sequence type (I.e. g., m., c. and n. for genomic, mitochondrial, coding DNA, and non-coding DNA.)

So far, I basically only have...

if chromosome in [1-23, X, Y] and position is integer:
    then "g"
if chromosome==M: 
    then "m" 
if [referencing annotation transcripts]: 
    then "c"

This seems certainly incomplete, and oversimplified :/