From: Mathias W. <wa...@eb...> - 2021-08-17 16:49:31
|
Dear PSI-MS CV community, I am writing on behalf of the PSI-QC working group, as we want to get our QC-CV 'into the fold'. Before that however, we thought it'd be best to start a conversation on how. We have been developing a separate CV initially to have some freedoms on what/how things work out best for the mzQC format, which relies heavily on CV control. Our CV holds the metric definitions, relations and few additional 'organisational' terms, so far <300 entries, for which we chose the QC prefix (naturally). We ditched the `xref: value-type` declarations in favour of defining term values with full CV terms and relations, which works nicely with the format. The mzQC format is, unlike most PSI formats, JSON based. One of the goals in the design was to bring the format closer to the last miles of analysis (R/pandas/web) but also as something that works between any steps of a MS based study workflow (lightweight with high automation potential). If you want to know more, check out our specification document draft (https://hupo-psi.github.io/mzQC/). For mzQC, we have to cover a vast range of QC/MS related topics in the CV. For that, we've devised a set of templates and rules for adding new term/metrics via GitHub issues. To keep better track of the entries, Nils created a GitHub workflow for obo-to-markup conversion to enhance browsability. QC-CV is not standalone, we base as much as possible on existing definitions, so we imported few other CVs: UO, OBI, STATO (,NCIT) and PSI-MS of course. So, overall I think QC-CV would fit in nicely, either as big PR or piece-wise. As for the merge, maybe we can follow the same playbook as the PEFF integration did? What do you think? TL;DR: mzQC wants to merge their CV into PSI-MS. But before opening a gargantuan PR, we should talk. Best, -- Mathias Walzer European Bioinformatics Institute (EMBL-EBI) Wellcome Trust Genome Campus, Hinxton, Cambridge, UK Office: +44 (0)1223 494 2610 E-mail: wa...@eb... |