The workflow for quality control of library spectra I have in mind would look something like this:
Complete annotation:
Harmonize metadata fields.
Correct wrongly entered fields. e.g. adduct is often added after the compound name.
Derive missing fields from other fields. E.g parent mass from adduct + precursor mz or charge from adduct. Or smiles from compound name (search on pubchem)
Check if metadata is correct:
Smiles should be a real smiles.
Smiles mass should match parent mass.
Charge, adduct and ion mode should all match.
Precursor mz, adduct and precursor mz should match.
Check if fragments would match smiles. For example #1 and #7
Automatically suggest corrections for common mistakes:
Suggest adduct replacement, based on precursor mz and parent mass.
Remove ions from salts.
The user should get a clear overview of the changed fields. The completions can probably just be highlighted with colour, while the automatic corrections should get a confirmation by the user.
The workflow for quality control of library spectra I have in mind would look something like this:
Complete annotation:
Check if metadata is correct:
Automatically suggest corrections for common mistakes:
The user should get a clear overview of the changed fields. The completions can probably just be highlighted with colour, while the automatic corrections should get a confirmation by the user.