molecular_mass of PubChemPureSubstance

I found that the molecular_mass for a pubchemPureSubstance is different from the Pubchem website. Why is it? Is this due to some unit conversion in the background of NOMAD? there are also unnecessarily many digits, when the error from unit conversion seems much larger. For an example see e.g. ethene:

Whereas on the PubChem website the mass is 30.07 g/mol.

changed the description

@hnaesstroem have you noticed this before? @himanel1 any idea what the problem could be?

@jschumann: Looks like we store either the exact mass or monoisotopic mass that are also shown on the PubChem page. I don't know how these differ from molecular weight. @hnaesstroem probably knows more. Also probably worth noting this in the quantity description if not already there.

Hrmm, that is strange. Did you check what the API response for that chemical is?

Ah, yes @himanel1 you are correct, this is the exact or monoisotopic mass. It is not only for the chemical above, but for all. The difference is noticable for carbon containing compounds, because of the relatively high 13C isotope concentration in naturally occuring compounds. So then the question is, do we want the exact mass @hnaesstroem ? I think molecular mass makes more sense to store here, right?

I think that is why I am using the unit Dalton. I'm not sure we have much choice if that is what PubChem returns.

oh, I see. it is in the basesection code, that you save the ExactMass. In this case we should correct the description. But I will double check if they not also return the molecular mass

Well, technically "molecular mass" is the mass of one single molecule measured in Dalton. en.wikipedia.org/wiki/Molecular_mass

I believe you're thinking about "molar mass" en.wikipedia.org/wiki/Molar_mass

Seems there is also a MolecularWeight endpoint in the PubChem API: https://pubchem.ncbi.nlm.nih.gov/rest/pug/compound/cid/6324/property/MolecularWeight

But I wouldn't assign that to the molecular_mass quantity in NOMAD. One option would be to add a molar_mass = Quantity(unit='kg/mol') quantity to the PureSubstanceSection and populate that with the PubChem MolecularWeight endpoint.

Here is a list of all properties in case there is something else you feel like adding:

<xs:element name="PropertyTable">
<xs:complexType>
<xs:sequence>
<xs:element name="Properties" maxOccurs="unbounded">
<xs:complexType>
<xs:sequence>
<xs:element name="CID" type="xs:int"/>
<xs:element name="MolecularFormula" type="xs:string" minOccurs="0"/>
<xs:element name="MolecularWeight" type="xs:string" minOccurs="0"/>
<xs:element name="CanonicalSMILES" type="xs:string" minOccurs="0"/>
<xs:element name="IsomericSMILES" type="xs:string" minOccurs="0"/>
<xs:element name="InChI" type="xs:string" minOccurs="0"/>
<xs:element name="InChIKey" type="xs:string" minOccurs="0"/>
<xs:element name="IUPACName" type="xs:string" minOccurs="0"/>
<xs:element name="XLogP" type="xs:double" minOccurs="0"/>
<xs:element name="ExactMass" type="xs:string" minOccurs="0"/>
<xs:element name="MonoisotopicMass" type="xs:string" minOccurs="0"/>
<xs:element name="TPSA" type="xs:double" minOccurs="0"/>
<xs:element name="Complexity" type="xs:int" minOccurs="0"/>
<xs:element name="Charge" type="xs:int" minOccurs="0"/>
<xs:element name="HBondDonorCount" type="xs:int" minOccurs="0"/>
<xs:element name="HBondAcceptorCount" type="xs:int" minOccurs="0"/>
<xs:element name="RotatableBondCount" type="xs:int" minOccurs="0"/>
<xs:element name="HeavyAtomCount" type="xs:int" minOccurs="0"/>
<xs:element name="IsotopeAtomCount" type="xs:int" minOccurs="0"/>
<xs:element name="AtomStereoCount" type="xs:int" minOccurs="0"/>
<xs:element name="DefinedAtomStereoCount" type="xs:int" minOccurs="0"/>
<xs:element name="UndefinedAtomStereoCount" type="xs:int" minOccurs="0"/>
<xs:element name="BondStereoCount" type="xs:int" minOccurs="0"/>
<xs:element name="DefinedBondStereoCount" type="xs:int" minOccurs="0"/>
<xs:element name="UndefinedBondStereoCount" type="xs:int" minOccurs="0"/>
<xs:element name="CovalentUnitCount" type="xs:int" minOccurs="0"/>
<xs:element name="Volume3D" type="xs:double" minOccurs="0"/>
<xs:element name="XStericQuadrupole3D" type="xs:double" minOccurs="0"/>
<xs:element name="YStericQuadrupole3D" type="xs:double" minOccurs="0"/>
<xs:element name="ZStericQuadrupole3D" type="xs:double" minOccurs="0"/>
<xs:element name="FeatureCount3D" type="xs:int" minOccurs="0"/>
<xs:element name="FeatureAcceptorCount3D" type="xs:int" minOccurs="0"/>
<xs:element name="FeatureDonorCount3D" type="xs:int" minOccurs="0"/>
<xs:element name="FeatureAnionCount3D" type="xs:int" minOccurs="0"/>
<xs:element name="FeatureCationCount3D" type="xs:int" minOccurs="0"/>
<xs:element name="FeatureRingCount3D" type="xs:int" minOccurs="0"/>
<xs:element name="FeatureHydrophobeCount3D" type="xs:int" minOccurs="0"/>
<xs:element name="ConformerModelRMSD3D" type="xs:double" minOccurs="0"/>
<xs:element name="EffectiveRotorCount3D" type="xs:double" minOccurs="0"/>
<xs:element name="ConformerCount3D" type="xs:int" minOccurs="0"/>
<xs:element name="Fingerprint2D" type="xs:base64Binary" minOccurs="0"/>
<xs:element name="Title" type="xs:string" minOccurs="0"/>
<xs:element name="PatentCount" type="xs:int" minOccurs="0"/>
<xs:element name="PatentFamilyCount" type="xs:int" minOccurs="0"/>
<xs:element name="LiteratureCount" type="xs:int" minOccurs="0"/>
</xs:sequence>
</xs:complexType>
</xs:element>
</xs:sequence>
</xs:complexType>
</xs:element>

Not sure about the difference between MonoisotopicMass and ExactMass... For ethane: ExactMass = 30.0469501914 MonoisotopicMass = 30.0469501914 MolecularWeight = 0.07

Thanks Hampus, I think it would be good to add molar_mass, and add to the description to molecular_mass that this is the monoisotopic mass. But I dont think more properties are needed at the moment. I think exact and monoisotopic are the same? I agree it is strange that they list the two.

from the pubchem website:

Exact Mass

The exact mass of an isotopic species is obtained by summing the masses of the individual isotopes of the molecule.
Monoisotopic Mass

The monoisotopic mass is the sum of the masses of the atoms in a molecule using the unbound, ground-state, rest mass of the principal (most abundant) isotope for each element instead of the isotopic average mass.

I guess ExactMass is then the monoisotopic mass unless the CID is referring to a specific isotope. Let me check deuterium to confirm!

Hrmm, no that doesn't seem to be the case for Deuterium:

<PropertyTable xs:schemaLocation="http://pubchem.ncbi.nlm.nih.gov/pug_rest https://pubchem.ncbi.nlm.nih.gov/pug_rest/pug_rest.xsd">
<script/>
<Properties>
<CID>24523</CID>
<MolecularWeight>4.0282035557</MolecularWeight>
<ExactMass>4.0282035557</ExactMass>
<MonoisotopicMass>4.0282035557</MonoisotopicMass>
</Properties>
</PropertyTable>

And not for heavy water either:

<PropertyTable xs:schemaLocation="http://pubchem.ncbi.nlm.nih.gov/pug_rest https://pubchem.ncbi.nlm.nih.gov/pug_rest/pug_rest.xsd">
<script/>
<Properties>
<CID>24602</CID>
<MolecularWeight>20.028</MolecularWeight>
<ExactMass>20.023118175</ExactMass>
<MonoisotopicMass>20.023118175</MonoisotopicMass>
<Title>Deuterium Oxide</Title>
</Properties>
</PropertyTable>

Normal water for reference:

<PropertyTable xs:schemaLocation="http://pubchem.ncbi.nlm.nih.gov/pug_rest https://pubchem.ncbi.nlm.nih.gov/pug_rest/pug_rest.xsd">
<script/>
<Properties>
<CID>962</CID>
<MolecularWeight>18.015</MolecularWeight>
<ExactMass>18.010564683</ExactMass>
<MonoisotopicMass>18.010564683</MonoisotopicMass>
<Title>Water</Title>
</Properties>
</PropertyTable>

created branch 2141-molecular_mass-of-pubchempuresubstance to address this issue

mentioned in merge request !2118 (merged)

mentioned in commit 92378802

closed with merge request !2118 (merged)