Yanjing Li

Why treat water in protein as glycine

2

In your protein.py (https://github.com/DeepGraphLearning/torchdrug/blob/a959f68f0c19f368be9e380f5a587de6970b3c67/torchdrug/data/protein.py#L1053), when the residue type is unknown (e.g. HOH: water), the residue will be treated as glycine. Will this affect the overall information of the protein? Why...

How to handle multiple chains in protein structure

1

Hi, when dealing with a protein structure with several chains, how does `Protein` class handle that? Does it connect all chains eg. a concatenated residue types list?

Question about self.task

1

Hi, in your implementation of GeneOntology dataset (https://github.com/DeepGraphLearning/torchdrug/blob/a959f68f0c19f368be9e380f5a587de6970b3c67/torchdrug/datasets/gene_ontology.py#L84), what is the `self.task` in line 144 and why we need it?

Hi, I recently encountered a problem when trying to run GearNet in TorchDrug on Google colab. I didn't use `pip install` or `conda install` to install TorchDrug. I use `git...

Solution when alpha carbon coordinate is missing

1

Hi, since GearNet models each residue in the protein by the C_alpha coordinates, how do you handle the situation when the coordinate for alpha carbon is missing?

Add dataset and finetune scripts for FOLD, ECSingle, GO, and ECMultiple for model GVP, CDConv, ProNet, and GearNet

libmolgrid issues about stratifying receptors

1

Hi authors, I have an issue using the `stratify` functions in `ExampleProvider`. I tried two ways : 1. `train_samples = molgrid.ExampleProvider(ligmolcache=args.trligte, recmolcache=args.trrecte, shuffle=True, default_batch_size=args.batch_size, iteration_scheme=molgrid.IterationScheme.SmallEpoch, balanced=True, stratify_pos=3, stratify_step=1, stratify_max=6, stratify_min=0)`...

Question about the RES task

1

Hi, I have a question about the RES task. For each example, the subunit indices contain atoms and residues from different chains of the protein. Do you just treat them...

Missing MajorPTMs train CSV file

1

Hi authors, I think the proteinBert benchmark data folder misses the train CSV for phophositePTMs (https://github.com/nadavbra/protein_bert/tree/master/protein_benchmarks). Can you help check this? Thanks.