There may be an issue with the order of the B-Factor

[ChenPy00]

https://github.com/NVIDIA-Digital-Bio/proteina/blob/main/graphein_utils/graphein_utils.py#L901

I noticed that directly using groupby to extract the b_factor in non-sequential output. To preserve the original sequence of the data, a viable approach is to add the sort=False parameter, like this: df.groupby("residue_id", sort=False)

Just wanted to share this small observation—it’s a minor detail and won’t impact the correctness of subsequent workflows, but I thought it might be helpful for maintaining data order consistency if needed.