Please cite the data you used

[cthoyt]

I'm surprised you have published this work and mention Protein Data Bank accession numbers, but make no mention of the Protein Data Bank (PDB) as the source of the data, nor cite to it.

For context, the PDB is the resource that collects, curates, and archives protein crystal structures and assigns them identifiers like 8OZC (e.g. that appears in the label of Figure 3A of your preprint).

This database is a massive international effort that relied on the primary work of tens of thousands of people over decades and the tireless biocuration efforts of the PDB teams in the same period. Here are the pages from three of the outputs in North America, Europe, and Asia which each have instructions on how to cite their most recent papers:

• https://www.ebi.ac.uk/pdbe/citing-us
• https://www.rcsb.org/pages/policies
• https://www.wwpdb.org/about/cite-us

I think it's only fair to acknowledge these efforts both on your repository on GitHub and in your manuscript.

[mbautistamartin]

Thanks for catching this, totally agreed on the value of PDB and we are updating our acknowledgement in the repo in the next few hours. The manuscript will also be updating in the following days to reflect this as well.

[mbautistamartin]

I have updated the acknowledgements in the repo and the manuscript will be updated on Arxiv by Tuesday. Thanks again for flagging this, I agree it's important and I'm happy to have mended this.

[yuyangw]

I have updated the arXiv preprint (https://arxiv.org/abs/2509.18480) to include references to PDB data. Thank you for catching this.

[cthoyt]

Thank you for taking the first step!

However, in the same sentence where you mention and cite PDB as a source, you mention SwissProt without citing it. Much like PDB, this is a large-scale curation effort that is worth acknowledging by minimally giving a citation to one or many of the ones listed on https://www.uniprot.org/help/publications

Btw, citation isn't the only thing. If Apple is using these data, maybe it can consider making a concrete monetary grant to PDB and/or SwissProt.

[drumyerscough]

Reopening this thread because on a related note:

1. The preprint contains some odd citation choices, eg "SimpleFold strongly departs from the design choices in previous work (Chakravarty and Porter, 2022)" when it might make more sense to reference the actual AlphaFold2 paper by Jumper et al.

2. Additionally, in Figure 10, I think the PDB ID for the leftmost protein is incorrect (just from looking at the monomer in the PDB entry). I think that this protein is actually calmodulin or a homolog.