Fields to decide for the sdrf metabolomics; [REQUIRED, RECOMMENDED, OPTIONAL]

[ypriverol]

During the PR from @mmattano #680 we detected multiple fields that we need to decide if they will be required, recommended or optional:

• scan polarity
• label
• fraction
• separation method
• metabolite assignment file

To be discussed.

[nilshoffmann]

scan polarity should be required and defined. positive is often the standard, but in lipidomics, we regularly use negative mode for richer fragmentation or even both modes for complementary information.
label could be optional. Usually only used in (semi)quantitative studies. Is that a spiked in internal standard with heavy isotopes? Would we require more details here? Example in lipidomics would be a lipidname + D7 for a lipid deuterated with seven Deuterium (label positions may not be available). In quantitative lipidomics, there is typically at least one labeled, lipid-class-specific internal standard added to each sample. Regularly, there can be multiple labeled internal standards added to each sample to quantify different lipid classes in one go. Usually, these labeled standards are sourced from commercially available labeled standard mixes.
fraction I cannot really comment on this. Not something I have seen in routine use in (targeted) lipidomics.
separation method should be required, but should then allow value for direct infusion, nanoflow injection etc.
metabolite assignment file should be required and link to either a file following the format of the MetaboLights MAF file or to another format like mzTab-M, which reports details down to (grouped) features, identification and quantification methods. In the case of MetaboLights, you can have one MAF for positive and one for negative mode identifications.

Can we require certain fields for different flavors of SDRF? E.g. clinical targeted metabolomics would require the label field to be present and filled?