search

enpkg / enpkg_graph_builder

ENPKG graph builder
GNU General Public License v3.0
1 stars 0 forks source link

About enpkg:SpectralPair #2

Closed mpagni12 closed 1 year ago

mpagni12 commented 1 year ago

(1) The value of enpkg:has_mass_difference is a float. Its sign depends on which spectrum is considered first in the difference of M/Z. This should be made explicit, for example by considering two new predicates:

enpkg:has_member_1 rdfs:subPropertyOf enpkg:has_member .
enpkg:has_member_2 rdfs:subPropertyOf enpkg:has_member .

Relying on the semantics of the IRI alone is not a good idea.

(2) I am not convinced that the distinction between enpkg:LFpair and enpkg:CSpair is very useful. At some point we may want to compare consensus and individual spectra as well, which would require defining a new subClass to be coherent. Maybe it is not necessary?

ArnaudGaudry commented 1 year ago

1) The mass difference should be only positive values (3a6ae8e) 2) This distinction was removed in 3a6ae8e

mpagni12 commented 1 year ago

The latest release perfectly address the above issue for enpkg:LCMSFeature. Indeed the parent mass of member_1 is always greater or equal to the one of member_2. Nice.

However, this is not yet implemented for enpkgdemo:GNPSConsensusSpectrum.

In addition, GNPSConsensusSpectrum might be considered as part of enpkg namespace (not enpkgdemo), for practical reason.