- Webserver and API synchronized as the user updates the data on the web interface; NMR-STAR file generated at the API server each time its synchronized
- Method to validate the dictionary update; should not break the existing system
- Human or automated testing framework to check the interface (UI) works correctly
- Maintain audit/revision record for the dictionary. It should be in the dictionary
- Allow interface/editing entries outside of web interface (command line deposition)
- Allow decoupling dictionary version from web interface
- Allow decoupling web server from API development
- Instant data validation
- Data consistency checking
- ORCID ID validation
- Email validation
- Depositor will be able to modify previously entered data in later stage
- Interface with ETS
- Interface with validation system
- Distribution to PDBj-BMRB
- Interface with wwPDB
- Creating development, testing, and production versions
- Avoid wwPDB problem where the system cannot be tested, because valid PDB codes are generated?
- In what formats will data be accepted
- Will depositions be locked and not editable after submission?
- Will depositors interact through annotators after completing depositions as is now?
- Deposition security (limit access to valid depositor or BMRB staff)
- Ease of access by BMRB staff to trouble shoot issues for users
- Industrial versus academic depositors
- User has ability to jump from one section to any other section of the deposition including data upload. The user is not forced to enter data in a pre-defined order
- Navigation tool is available that describes the different sections of the deposition and ability to jump from section to section
- Sections with mandatory fields are clearly distinguished from sections that have no non-mandatory fields
- Mandatory fields are clearly defined
- Mandatory fields that have been filled in are clearly distinguished from those that have not been completed
- Entry hold choices
- Fields that have fixed enumerations
- Fields with open enumerations
- Mandatory fields that are usually applicable, but in some cases may not have true values (pH in pure organic solvents)
- Ability to import data from BMRB database (e.g. chem comp information for a ligand or non-standard residue)
- What kinds of data to accept?
- NMR-STAR dictionary
- Conditional mandatory linkages (files?)
- Information field linkages - fields that can use a pull down list populated from other sections
- Somewhat defined by the BMRB entries listed below for testing
- Single chain protein only (<entry number?>)
- Single chain protein with one organic ligand (<entry number?>)
- Single chain protein with metal ligand and organic ligand (<entry number?>)
- Protein homodimer (<entry number?>)
- Protein heterodimer (<entry number?>)
- Protein polynucleic acid complex (<entry number?>)
- RNA monomer (<entry number?>)
- RNA duplex (<entry number?>)
- RNA/DNA hybrid monomer (<entry number?>)
- DNA duplex (<entry number?>)
- RNA monomer with ligand (<entry number?>)
- Chemical shifts (<entry number?>)
- Coupling constants (<entry number?>)
- Relaxation data (<entry number?>)
- Students use data from entries to populate deposition
- each student uses 4-5 entries and the entries used rotates to keep fresh eyes looking for bugs
- Students also 'monkey' test fields and sections of the deposition not commonly used (just type in junk to quickly test functionality)
- Senior staff also test system
- All issues discovered entered into tracking system
- Issues retested by person reporting the issue, as fixes are reported
- Test over and over again as the system evolves and is updated
- NMRFAM personnel asked to review and provide feedback on UI
- As system nears completion, ask a few outside users (BMRB curators?) to test system
- Measure time user takes between starting a deposition and completing the deposition