-
Notifications
You must be signed in to change notification settings - Fork 2
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
added new translations by Lex #21
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Error: leading Whitespace Check
- count: 457
- error_code: E2
Error: trailing Whitespace Check
- count: 882
- error_code: E3
Example:
en nl Hib vaccine (HP:0410305 rdfs:label Partial absence of specific antibody response to Haemophilus influenzae type b Gedeeltelijke afwezigheid van specifieke antilichaamrespons op Haemophilus influenzae type b OFFICIAL orcid:0000-0002-5648-2155 MD, Geneticist 2024-11-26
- Use strip() or trim() functions to ensure that no leading or trailing whitespace exist in any cell.
- somehow the last line in the spreadsheet got totally broken.
Not sure what you mean with the second point, for the first I need to import this in python, which currently I don't (it goes from google directly to babelon format), no problem but I need a few minutes to do it, which I will have tomorrow. |
This is not a legal row.. Where is the HP id? |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I will process it for now, but not that we have a schema specified for translator expertise: https://github.com/monarch-initiative/babelon/blob/ee210876516f54ca62ed8d33bbe175ddc5060375/src/babelon/schema/babelon.yaml#L19
Maybe do this in a future processing step!
I had no idea, sorry! Can we discuss adding more categories? I have a feeling a bioinformatician or a geneticist (if translating some term which is slightly beyond their field expertise) may neither fit in DOMAIN_EXPERT nor in LAYPERSON. We could have an intermediate category which is somewhat in between...? |
Of course! Any suggestion? Shall we just add those domain specific categories? I don't mind so much. Better of course would be something like "technical specialist", "domain informatician" (if that existed) Md imo is clearly domain expert? |
Well, I can think of 2 things. 1) no constraint, just let people write what they prefer (maybe some constraint like using latin characters and maybe commas, dashes and dots) and 2) add "technical specialist" for anyone who is not a layperson but also not a domain expert. |
for the unconstrained case i would prefer using an open ended field like comment. we want to be able to make statistics across expert types, and an pure string field is too much work to clean for generating stats IMO. Technical specialist it is! Feel free to add to the babelon schema in PR, I will make a new release and you can add it here |
@leokim-l I changed the formatting a bit so its easier reviewing this in the future. Can you check out how the file looks like when you get a chance? |
Unfortunately something went wrong here, I see translations in the google doc that are not in here... |
No description provided.