The editors' position on open science is quite noteworthy:
"As for open science, we have set the requirement for data publishing high, and we want to see original data sets in the supplementary information or linked to other journals dedicated to the publication of data."
If you are on an editorial board for a journal that lacks an open science statement like this, you might find this JPA approach useful as a template to suggest to your editor some updates to journal policy.
Similarly, when we do peer reviews we can encourage authors to deposit the data behind figures and tables in trustworthy repositories, and add a DOI to the paper to link to the dataset, nothing that this is a requirement for some journals.
best,
Ben