search

biocore / LabControl

lab manager for plate maps and sequence flows
BSD 3-Clause "New" or "Revised" License
2 stars 15 forks source link

Instrument_model is a standard vocabulary in EBI #430

Open ackermag opened 5 years ago

ackermag commented 5 years ago

The developer group agreed to pre-populate with "Illumina MiSeq" since that is the most frequently used instrument.

charles-cowart commented 5 years ago

@ackermag I might need to clarify where in the UI this is occurring.

charles-cowart commented 5 years ago

To clarify, this is also a column in one of the prep information files - from my notes, it presently reads 'MiSeq'

charles-cowart commented 5 years ago

High priority as it reflects what wet-lab has deemed proper output.

charles-cowart commented 5 years ago

In samples and controls files this column is named 'INSTRUMENT_MODEL'.

antgonza commented 5 years ago

In case it helps: https://github.com/biocore/qiita/blob/master/qiita_ware/ebi.py#L93

charles-cowart commented 5 years ago

This will also require changes to the EXPERIMENTAL_SAMPLES_PREP_EXAMPLE expected data string in unit tests. This should be done after https://github.com/jdereus/labman/issues/429 is merged.

charles-cowart commented 5 years ago

Thanks Antonio!

On Tue, Apr 16, 2019 at 11:57 AM Antonio Gonzalez notifications@github.com wrote:

In case it helps: https://github.com/biocore/qiita/blob/master/qiita_ware/ebi.py#L93

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/jdereus/labman/issues/430#issuecomment-483800913, or mute the thread https://github.com/notifications/unsubscribe-auth/AKFU7EZMQ25BHBIQ6L5NBL3PQYORJANCNFSM4HDN6DIQ .

charles-cowart commented 5 years ago

Consider doing in conjunction w/ https://github.com/jdereus/labman/issues/431, so that the expected data only has to be modified once.

AmandaBirmingham commented 5 years ago

FYI @wasade This is SOOOO easy (and will save Gail having to do a manual search and replace) that I am pulling it into the 0.1.0 milestone. I am already mucking around in the prep sheets for #431 and #445 anyway :)

AmandaBirmingham commented 5 years ago

@ackermag @antgonza I am ready to make this change but need to know exactly what the correct contents of the prep sheet INSTRUMENT_MODEL should be for this sequencer: Above from @ackermag (https://github.com/jdereus/labman/issues/430#issue-428984513) I have Illumina MiSeq but from @antgonza (https://github.com/biocore/qiita/blob/master/qiita_ware/ebi.py#L93) I have ILLUMINA MISEQ. Which should I put in the prep sheet?

Note that the full list of sequencer instrument types that LabControl currently recognizes is 'HiSeq4000', 'MiniSeq', 'NextSeq', 'HiSeq3000', 'HiSeq2500', 'HiSeq1500', 'MiSeq', 'NovaSeq'. It seems unlikely that MiSeq is the only one of these that should be changed to match the EBI standards, and now would be a good time to change any of them that need changed. Please provide the exact string you want in the prep file for each sequencer for which a change is needed!

antgonza commented 5 years ago

We just made that column and a bunch other a requirement to make any artifact private/public in Qiita and the final official values are going in this PR: https://github.com/biocore/qiita/pull/2890/files; note that it still needs to be reviewed by @ackermag and @charles-cowart.

ackermag commented 5 years ago

Are we not including the Oxford Nanopore machines - Minion etc?

The list Antonio has pointed to is the full set of all machines accepted by Qiita but some are not relevant to LabControl and will never be (LS454 machines and Ion Torrent). I would specify the following: Illumina HiSeq 1000', 'Illumina HiSeq 1500', 'Illumina HiSeq 2000', 'Illumina HiSeq 2500', 'Illumina HiSeq 3000', 'Illumina HiSeq 4000', 'Illumina MiSeq', 'Illumina MiniSeq', 'Illumina NovaSeq 6000', 'NextSeq 500', 'NextSeq 550', 'MinION', 'GridION', 'PromethION',

But Greg needs to sign off on these options

On Fri, May 3, 2019 at 3:40 PM Antonio Gonzalez notifications@github.com wrote:

We just made that column and a bunch other a requirement to make any artifact private/public in Qiita and the final official values are going in this PR: https://github.com/biocore/qiita/pull/2890/files; note that it still needs to be reviewed by @ackermag https://github.com/ackermag and @charles-cowart https://github.com/charles-cowart.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/jdereus/labman/issues/430#issuecomment-489261833, or mute the thread https://github.com/notifications/unsubscribe-auth/AAPL3ADYN2PXBMH2YDZIXFTPTS5NJANCNFSM4HDN6DIQ .

-- Gail Ackermann Knight Lab UCSD glackermann@ucsd.edu ackermag@ucsd.edu

AmandaBirmingham commented 5 years ago

@ackermag @antgonza Why is Gail's list mixed case while Antonio's is all upper-case? (like, is there some additional processing between the prep sheet and the EBI upload from Qiita that ucases everything? Or do I need to worry about the discrepancy? :)

ackermag commented 5 years ago

My list is exactly as EBI requires the text to be inserted - it is case-specific for how it should be inserted into prep sheet. e.g. Illumina MiSeq is fine but Illumina Miseq will be rejected.

AmandaBirmingham commented 5 years ago

Eww, not quite as easy as I thought because of https://github.com/jdereus/labman/issues/266

AmandaBirmingham commented 5 years ago

Moved back out of 0.1.0 milestone because of linkage to #266 above and need (stated above by Gail) to get Greg's approval on the new values.

AmandaBirmingham commented 5 years ago

See https://github.com/jdereus/labman/issues/507#issue-442364742 re related issues to be fixed at same time as this.