About annotation for peptide mixtures

[daichengxin]

Hi All,In the LC-MS Analysis of Isoaspartylated Peptides part of the PXD006112 project, the experiment uses a mixture of 12 peptides to conduct the experiment, so Organization、 Organization part, etc. will not be applicable, and some unique information such as peptide quality may need to be added. Maybe we can learn from the key-value method of the spiked-in experiment。

[daichengxin]

Source Name	Peptide	Peptide	data file
Sample1	CT=peptide;Mass=976.448;PS=AGFAGDDAPR	CT=peptide;Mass=1265.612;PS=DGNGYISAAELR	20140919_QE6+_LC5_CY_SA_IsoAsp_40min_12mix_01

[levitsky]

Hi, could you please specify the question or suggestion?

I see you've added peptide molecular masses in the annotation and are talking about other keys, but in the example table only masses are listed. In this case, however, the mass can be trivially calculated from the sequence, so I'm not sure I understand the benefit of adding it.

[daichengxin]

Sorry. First of all, the experiment uses a mixture of 12 synthetic peptides, so organism, organism part, cell type, etc. will no longer be applicable. I am not sure if it can pass sdrf-validation. Then for peptide 11 and peptide 12, where the isoaspartic acid residues are underlined. I am not sure if using underscores in annotation is a good option to express the isomer relationship of amino acids.Finally, I think both Compound vendor (CV) and Compound specification URI keys should be applicable in this case. For this article, CV=Thermo Scientific.

Thanks!!

[levitsky]

Ah, sorry! I was thinking you about the annotation scheme for spiked compounds, but if in this case the sample itself is a relatively simple peptide mixture, then it's a different story.

We have some examples with synthetic proteomes already annotated. Organism and some other things are still mandatory, but can be set to "not applicable". There is also characteristics[synthetic peptide] that should be added. Arguably this scheme applies to your case.

Exhaustively annotating every component of the sample might be too much. Perhaps a comment with some specification URI is all that's needed here. Twelve peptides per file, including non-standard amino acids, seem like a lot of work to fit into the format. Do you think it's worth it?