Inconsistent receptor-ligand order

[ktravaglini]

Thank you for curating a list of protein receptor-ligand interactions, very helpful!

I noticed that the majority of interactions are listed as LIGAND_RECEPTOR in the database, but every now and then you run across the opposite.

Ex: Most of the Wnt interactions are WNT_FZD, but a few like FZD1_WNT3A are reversed.

Is it possible to make these consistent in a future release? Or Is there a simple way to pass a custom interaction list within the CLI?

Kyle

[mvento]

Hi @ktravaglini,

We will try to fix it in the future, sorry for the inconvenience.

Thanks for using CellPhoneDB

[aditisk]

Hi @mvento, is there any way yet for users to use a custom list of interactions ? @ktravaglini were you able to use a custom list yet or is that feature still upcoming ?

Thank you.

[mvento]

Hi @aditisk,

Now you can use your custom list when you generate your custom database.

Please, update CellPhoneDB version and check the documentation:

https://github.com/Teichlab/cellphonedb#examples-for-user-specific-custom-database

Best