FoldCompDataset calls graphein.protein.tensor.Protein with incorrect __init__ arguments

[mawright]

Hi y'all, love the library. It's been a huge help for me so far. I wanted to make use of the foldcomp integration but it isn't creating the Protein data objects correctly.

Describe the bug
This part here in fc_to_pyg:

graphein/graphein/ml/datasets/foldcomp_dataset.py

Lines 297 to 305 in 47d9dde

	return Protein(
	coords=torch.from_numpy(coords).float(),
	residues=res,
	residue_id=[f"A:{m}:{str(n)}" for m, n in zip(res, res_num)],
	chains=torch.zeros(len(res)),
	residue_type=residue_type.long(),
	b_factor=torch.from_numpy(b_factor).float(),
	id=name,
	)

tries to construct a Protein object, but Protein's init method looks like

graphein/graphein/protein/tensor/data.py

Lines 134 to 136 in 47d9dde

	def __init__(
	self, atom_list: List[str] = PROTEIN_ATOMS, fill_value: float = 1e-5
	) -> None:

so trying to get proteins out of FoldCompDataset always errors with TypeError: __init__() got an unexpected keyword argument 'coords'

To Reproduce
Steps to reproduce the behavior:

1. Run the first three cells of foldcomp.ipynb.

Expected behavior
FoldCompDataset to iterate through Protein objects

Desktop (please complete the following information):

• OS: Ubuntu 22.04
• Python Version: 3.9
• Graphein Version [e.g. 22] & how it was installed: 47d9dde, from source

[a-r-j]

Hey @mawright thanks for pointing this out. I’d fixed this locally and forgotten to update this. Will merge as soon as tests pass. Apologies!

[mawright]

Thanks!