Question about motif scaffolding benchmarking

[LY-Xie]

Hi,

First of all thank you for sharing this wonderful work! I am currently trying to run the motif scaffolding benchmarking experiment but I have run into some problems:

1. How do you actually prepare the benchmarking dataset, i.e. download the motif pdbs etc. And it looks like the contig_string and segment_order have to be specified in the inference_motif.yaml file. Is there an automated script that can extract the contig_string and segment_order from the motif pdb file? Although 25 tasks are not a lot but manually changing them is still tedious nonetheless.
2. It looks like you hinted at formatting the segment_order for Scaffold_Lab. Are the results presented in your Supplementary Table 5 directly taken from the output of Scaffold_Lab?
3. It seems the segment_order specified in the config file is not actually passed on to the GenMotifDataset in motif_inference.py. So the resulting motif_info.csv has the default 'A' as the segment order instead of 'A;B;C' when I run python proteinfoundation/motif_inference.py. I thought it should be passed on from cfg when initializing the GenMotifDataset just like contig_string but I don't see it in the code. Am I understanding it wrong?

Thanks in advance,
Liyang