Compute DAG coverage score

[oolonek]

Displaying coverage percentage will help us to know what are the required efforts to complete the input dataset.

At the moment according to this test we have the 50 classes and 6 superclasses which are present in the DAG but absent of the training dataset.

Classes not in the dataset :

'Nonadrides', 'Fungal DPEs', 'Marine-bacterial DPEs', 'Breviane diterpenoids', 'Acyclic triterpenoids', 'Carotenoids (C40, ε-Ψ)', 'Carotenoids (C40, ε-ε)', 'Carotenoids (C40, β-Χ)', 'Carotenoids (C40, π-Χ)', 'Carotenoids (C45, ε-Ψ)', 'Polyprenol derivatives', 'Cephalotaxus alkaloids', 'Homoerythrina alkaloids', 'Ahp-containing cyclodepsipeptides', 'Monocyclic β-lactams', 'RiPPs-Amatoxins and Phallotoxins', 'RiPPs-Bottromycins', 'RiPPs-Cyanobactins', 'RiPPs-Lanthipeptides', 'RiPPs-Lasso peptides', 'RiPPs-Microcins', 'RiPPs-Thiopeptides', 'Halogenated hydrocarbons', 'Peptaibols', 'Miscellaneous alkaloids', 'Simple amide alkaloids', 'Miscellaneous polyketides', 'Simple tetramate alkaloids', 'Mitomycins', 'Orthosomycins', 'Phoslactomycins or Phosphazomycins', 'pteridine alkaloids', 'Phenalens', 'Naphthalenones', 'Prodigiosins', 'Ericamycins', 'Thiazole alkaloids', 'Azo and Azoxy alkaloids', 'Benzoquinones', 'Benzophenones', 'Boromycins', 'Merohemiterpenoids', 'Aplysiatoxins', 'Fungal cyclic polyketides (Miscellaneous)', 'Depsides', 'Depsidones', 'Usnic acid and derivatives', 'Duclauxin and derivatives', 'Furans', 'Phenanthrenes']

Superclasses not in the dataset :

['Diphenyl ethers (DPEs)', 'Miscellaneous alkaloids', 'Miscellaneous polyketides', 'Mitomycin derivatives', 'Serine alkaloids', 'Phenanthrenoids']