MaSC: Data Sharing Comments

From National Gallery Research Wiki
Jump to: navigation, search

Summary

Comments/requests/answers for Joe and David in relation to the MaSC: Data Sharing project.

This is an excellent and impressive interface!!!

Notes submitted by Ken, Chris, Ester & KJ organised and responded to by Jpadfield and Dpeggie 15:06, 13 March 2011 (UTC)

JOE

  • Check the data validation rules on the global lists. At the moment it will not let two entries with the same name. This should be changed to no duplicates of name for a given jcamp_id. DONE
It should now be possible to have entries with the same name but in different lists.
  • Make it possible to reorder the file lists by clicking on the table headings. DONE

Front Page

Ken

  • Please use full name: “Users’ Group for Mass Spectrometry and Chromatography” DONE
  • First line “This web page ... MaSC organisation” seems redundant – repeated below. Second line “If you have any questions ... visit the website” could perhaps go at the bottom of the page. DONE
This has been replaced with a general text saying: Mass Spectral data submission tool
  • NB: “questions” (pl.), omit “itself”. DONE

Ester

  • It would be useful to have more immediately accessible a description of the type of files that can be submitted. This comes quite far down in the long instruction page. Just a mention in the opening page would probably be enough. DONE

Chris

  • When asking users to "email MaSC" in the instructions or help "balloons," please include spectra@masgroup.org in parentheses. David and Joe can both have access to this account if they'd like. Also, the e-mail link for MaSC at the top right corner of the main page should link to spectra@mascgroup.org instead of David's address so that we don't have to change it later. DONE

Submission

Chris

  • Selecting a profile before uploading still requires that the inlet type, spectrometer, spectrometer type, and ionisation mode be filled in again anyway. Could this be taken care of in the user-defined profiles? DONE
If a jcamp file was uploaded that contained values in these fields that did not exactly 
correspond to those in the drop-down lists, the fields were originally blanked. 
We need to set it to automatically override drop-down variable values with those selected
in the profile.  This will be done but it should be noted that because the system accepts
jcamps a user could overwrite existing data in a jcamp file with the new profile data.
Existing values in uploaded files will now always be replaced by values defined in 
the selected profile.
  • Spectrometer: should have an "Other (user defined)" (KJ) choice. Otherwise, I'm afraid someone will just pick an instrument that's close to one listed and then forget to tell us what the real instrument is. DONE
Added "(Other: Defined in additional information)"
Please note as part of the MaSC checking process new spectrometers should be added to the
list and then this field should be updated.
  • Derivatization We would like to include a "Derivatization" field (see website). Please take out of ‘sampling procedure’ As it is not a standard JCAMP field, it should be preceded by a "$" and spelled with a "z". All other uses of derivatization can be spelled with "s". DONE
Not sure what you mean by "see website" but this has now been done, we think :-)

Ken

  • Is it possible to modify an existing profile and save it under a new name? Maybe I just couldn’t see how – but if not, is this a feature we could add? Might save errors where a number of similar methods are used (with different run times etc.) DONE
Added a "Save as New" and a "Delete" button as to the Edit profile form.
I have not added a "Delete" option for any thing else at this stage, as there are a 
variety of possible interdependencies with most of the values. However profiles are only 
used during the actual initial upload process, there is no long term connection between 
profiles and formatted files, so editing or deleting an old profile will not effect 
files that have already been processed. If this is not clear please let me know and I can 
add more information.
  • Title: let submitter write common/trivial name - the MaSC number could then be added upon conversion to the JCAMP. As it is, it may be a bit unclear to the user what's needed here. DAVID Todo
Perhaps what needs to be changed is the instructions for the user?
The actual "file" is created right after the upload stage and then is edited during the
submission/acceptance stage.  The number that currently appears in the name is related to
the unique ID number in the database list of files.  Using this number ensures that there
is no duplication of MaSC number. Would recommend keeping current process but amend 
information/instructions to ensure the process is clear to users, which we will attempt 
to do, please comment.
  • Original filename: This should refer to the full GCMS data file, not just the extracted spectrum (according to the original instructions the filename should "serve as a reference if the original data need to be accessed at a later date").
This field defaults to the name of the uploaded file.  However it has now been made 
editable so if the user has not named the upload file appropriately they can edit the 
value. 
  • Instrumental parameters: Some of us keep having a problem with this information getting cropped in the profile (128 characters seems to be the cut-off in the cases I counted), and having to re-enter the full information for each submission. DONE
    • KJ: If there must be a 128 char limit (but only then!)– please make field $Instrumental parameters 2
    • Chris: It would need the $ to indicate that it’s not a standard JCAMP field
The limit in the profile has now been increased to 512, this can be increased further if
required.
  • When I tried to edit text in a fixed field (just to try...) I was dropped out of the data entry page ...is this a glitch? DONE
There does not seem to be a specific bug here, or I was unable to replicate the problem. 
However the space-bar and return keys can effect the page if you are not actually editing
a value. If you can specifically replicate this problem please send us the details so we
can clarify the issue.
  • Acquisition range: should be entered by user, meaning the range at which the instrument was set for the acquisition of the current spectrum i.e. an instrument setting. DONE
    • (KJ) Change in the Info accordingly! Sorry for the conflicting info we gave you earlier.
This has also been added to the profile as it is related to the standard instrumental 
set-up 
  • Scan rate (KJ): give a drop-down menu for just the units - put in scan/sec instead and an option ‘other (user defined)’ make info: "scans/sec is proposed by MaSC" CONSIDERED
  • Resolution: ditto DONE
    • (KJ) amu and ‘other’ . David and Joe: please this should be the resolution of the instrument! DAVID Todo
Adding this type of sub drop down menu would required a fair amount of restructuring of
how the values are processed. We propose that the information is just amended to make 
MaSC's preferred units clear. This can possibly be revisited if required when we look at
version 2 of this system
  • Threshold: KJ) ‘units’ can be removed DONE
  • CAS name: David and Joe - is there any easy way to check this ...other than subscribing to SciFinder? I've been adding the CAS number (which can be found for free e.g. on the NIST site) but leaving the CAS name blank, as I’m not sure how to check it. DAVID to check
    • (Ester) I think this is a very relevant point. I also do not know how to check a CAS name even if I do have a number. We can change it to IUPAC nomenclature? This could be quite off putting for some of the natural products though.
    • (KJ) What do you suggest? Wikipedia gives stuff like http://chem.sis.nlm.nih.gov/chemidplus/– this seems to work?
  • Base peak: in one of my files this was incorrect (triphenylphosphate: 36 instead of 326) DAVID todo
    • (KJ) Is this a glitch?
May be for older uploaded files during testing of earlier version of the system, should 
be fine now.
  • "Submit/Cancel": I'm still not quite clear what the "Cancel" option means - when I tried this time the uploaded file was added to my list but I lost all information except the fixed fields (i.e. those pulled automatically from the data file) - is this a glitch? I'm presuming it should work by saving all entered info for completion/submission later. If so maybe "Save" or "Submit later" would be a better description. Done
    • (Ester) It would be useful to add a save/submit later button or at very least a warning that by pressing cancel you loose the added data.
Check to see if it is possible to allow "Save Edits" with out changing status and without 
doing data validation process.
A Save Changes" button has been added to the file form during when in uploaded or 
pending status.  Tool tips have also been added to buttons to make their function clearer.
These tool-tips can be edited if required.

KJ

  • Licence statement. Please put in: ‘By accepting this database, the user agrees to be bound by the terms of the MaSC user’s license. Any reference written/oral made to this file must include accreditation to BOTH the originating individual/institution and MaSC. Contributor agrees to be bound by the terms of the MaSC contributor’s license.’ DONE
The wording has been amended slightly as this is the statement that will appear in the 
actual file, not in a database. This type of process requires that the licences mentioned
are publicly accessible, so they will need to be posted on the MaSC website and linked
correctly. The clickable links in the info text also need to be updated. 
Are there two separate licences user's and contributor’s ?

MASC to provide links

By using this file, the user agrees to be bound by the terms of the MaSC user’s license. 
Any reference written/oral made to this file must include accreditation to BOTH the 
originating individual/institution and MaSC. Contributor agrees to be bound by the terms 
of the MaSC contributor’s license. The MaSC contributor’s license is available at 
http://www.mascgroup.org
  • Spectrometer type: put ‘designation of…’ in front of first sentence ‘if the correct…’ in info. DONE
  • DAVID to check other info boxes
  • Data processing: should be recoreder here : typo in info. DONE

Ester

  • Molecular weight/Nominal mass/ accurate mass: Is it possible to make the links in the information box live or highlighted. DONE
The Kovats' link was updated to a wikipedia link as the original link was no longer working.

JCAMP FILE

Chris

  • The use of "Submitted" and "Uploaded" is confusing since there is little distinction between the two terms in this context. I propose replacing "Submitted" with"Pending" indicating to the user that action needs to be/or is being taken by the MaSC Committee. Consequently, "Listall submitted files" should be "List all pending files". DONE
  • Also, the "Status" label for these types of files should be changed from "Submitted" to "Pending". DONE
  • Under "Status" - If an "accepted" file isunlocked and edited, it becomes an "uploaded" file and the date reverts to the date in the original JCAMP-DX file (which, by the way, should be and remain the date of sample analysis). The date under "Status" should be the date of "accepted","pending/submitted", etc. DONE
This will be fine, now the date of analysis is editable. The upload date has now been
separated for any LONG DATE values in the uploaded file and is purely used within the
database status field.
The date is now always the same as the status ie Uploaded, Pending, Accepted, if the file
is unlocked it reverts to its original uploaded status and uploaded date. There is no 
"unlocked" event/status 
  • "Date uploaded" should be changed to "Date collected"and defined as the date the spectrum was collected by instrumental analysis. DONE
Has been changed to "Date of Analysis" and is now an editable field and is automatically
pulled from an uploaded file if the field exists (LONG DATE). "Date Collected" was deemed
to be non specific as it could be taken to mean the date when masc collected the file. 
However, this display name is very easy to change as required.
  • It is wisest to refrain from using Greek letters and just use .alpha and .beta instead in case there are problems for some software conversion programs. DAVID todo
    • (KJ) Can this be put in the info menu?
  • Could the quotation marks in the uploaded JCAMP-DX files be automatically removed? See MaSC00008 as an example. DONE
This should be correct for all new files. The plan is to actually wipe all of the 
uploaded "test" files before live launch.

Ken

  • Lines in the JCAMP file should be no longer than 80 characters(including label): is it possible to have the software break longer entries into several lines automatically?DONE
  • Header information from the ASCII/txt file is added into the JCAMPfile as “$$extra”; “##FILELINE-0,1” etc... Is there anyreason these lines were included?? DONE
    • (KJ) Otherwise leave them out!
This information was kept as it was good not to throw away information that might be 
useful or important. But it has now be turned off and these extra lines of data will 
be dumped.
This extra data will now be discarded and not added when files are initially added. This
will only effect all new files uploaded.
  • For fields in which grey text was included (i.e. instructions to the submitter), and the field not filled out, I found the text was sometimes transferred into the JCAMP (although it’s possible this only happened for files that were “cancelled”, not “submitted”). DAVID Check
  • Sometimes I had difficulty navigating back out from viewing the JCAMP(it took several clicks), although sometimes it worked OK. DONE
This is probably an issue with going "back" to pages that were created with "POST"ed 
form data, this is a browser based issue and is common to all web-sites that use forms 
to submit data. The Back button can often cause issues in complex web-sites.
As a solution I have set the actual jcamp files to open up in a new tab or window, 
is this ok, or would you prefer it back the way it was before?
  • In the JCAMP files of several of my submitted files, the lines “.SCANNUMBER” and/or “$KOVATS INDEX” were lost (these fields were left empty, but the lines should still be preserved) - I didn’t check every JCAMP thoroughly so it’s possible there were others too. DONE
    • (Chris) All the lines from the standard MaSC JCAMP file need to be present in the submitted file even if they remain blank.
This seems to be working now, all of the fields should be added to the files once they 
have been submitted or made pending.

Ester

  • View spectrum. Peak labels are very small and maybe not all peaks need to be labeled. DONE
This is a simple graphing system designed to allow users to see if the uploaded data appears
as expected. It is currently possible to increase the font size and or hide labels under 
a certain value if requested.  
"Red" Zero values may occur due to the normalisation process (0-10000).
The ability to "zoom in" may be investigated later but not specifically 
as part of this project. 
  • Title correct intesity to intensity DONE