search

menchelab / datadivr-demo

Demo version of the DataDiVR 3-D network visualization software
MIT License
2 stars 0 forks source link

savegame to db #16

Open sepppirch opened 4 years ago

sepppirch commented 4 years ago

what parameters do we want to save?

ohaijen commented 4 years ago

Selections, definitely. Other stuff, maybe.

Keep in mind that, at least in theory, the data in the DB can update (for instance if we pull in new PPI or new GO data).

sepppirch commented 4 years ago

i think it would make sense to treat selections individually and not save them in the savegame struct but only revference them there. what do you think? concerning the update point: i guess we have to save them with entrezIDs then so we can re-associate them if the network changes

ohaijen commented 4 years ago

Without having thought too deeply about it, referencing selections and saving them elsewhere makes sense, with the caveat that we potentially run the risk of saving a lot of selections that are never used/useful, so maybe we should delete them once they get old.

EntrezIDs or internal IDs (maybe slightly preferred) should work for genes. I was more worried about saving the "perspective" since updating the PPI would likely update the layout as well.

On Fri, Nov 8, 2019 at 2:34 PM Sebastian Pirch notifications@github.com wrote:

i think it would make sense to treat selections individually and not save them in the savegame struct but only revference them there. what do you think? concerning the update point: i guess we have to save them with entrezIDs then so we can re-associate them if the network changes

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/menchelab/datadivr-demo/issues/16?email_source=notifications&email_token=AAA746476OQ4ZPASNRFNUHDQSVTEPA5CNFSM4JKDTAXKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEDR5AJI#issuecomment-551800869, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAA7462SF4K666VYP3EAHFLQSVTEPANCNFSM4JKDTAXA .

sepppirch commented 4 years ago

i see. dont worry too much about the perspective, it would also change if room calibration changes. nothing we can do there....

ohaijen commented 4 years ago

Maybe we can save something like save the identity and position of the closest two genes and return to a layout where those two genes are in those exact positions.

On Fri, Nov 8, 2019 at 2:49 PM Sebastian Pirch notifications@github.com wrote:

i see. dont worry too much about the perspective, it would also change if room calibration changes. nothing we can do there....

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/menchelab/datadivr-demo/issues/16?email_source=notifications&email_token=AAA746ZRYCJL6DQN42DR5HTQSVU6DA5CNFSM4JKDTAXKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEDSCPJQ#issuecomment-551823270, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAA74655XOFULITWTVRPSTLQSVU6DANCNFSM4JKDTAXA .

sepppirch commented 4 years ago

or we could come up with something like a pin object that marks an interesting area in a specific layout. preferably with ability to leave a note aswell. single genes should be easy to find anyway