Access to Multi-Domain Benchmark Dataset

[nwod-edispu]

Thanks for your great work!

I'd like to do some evaluation on the multi-domain benchmark which contains the AlphafoldDB protein as described in the paper. While I noticed that a file containing AFDB protein IDs is provided, I encountered difficulties downloading the entire AFDB dataset due to its substantial memory requirement. Would it be possible for you to share the specific protein structures included in your multi-domain benchmark directly?