jhakanchan15 / ppi_gnn Goto Github PK

!!!!!!!

I was trying to replicate your result. I followed README.md to set up this repo. However, when I try to run it I got a Runtime error as followed:

RuntimeError: The 'data' object was created by an older version of PyG. If this error occurred while loading an already existing dataset, remove the 'processed/' directory in the dataset's root folder and try again.

I beleive this error occur because of the DataLoader in the data_prepare.py. Any solution on how to solve this? Much appreciated.

I would like to replicate results mentioned in the paper. However, access to the original processed folder containing all 4,188 protein graphs is restricted and will require permission. I would greatly appreciate if you can assist and grant me access to the folder. Thank you very much.

I am trying to understand the pipeline for converting proteins to graphs, and I have found a difficulty in this line:

ftrs = np.load("../human_features/pdb_to_seqvec_dict.npy", allow_pickle=True)
I would like to know where did you get the file pdf_to seqvec_dict.npy? and what does it contains?
Appreciate any answer.

Hello there,

It would be greatly appreciated if you could also upload the environment file somewhere so that your results are reproducible.

Otherwise we have no idea how to run your model with which version of pytorch etc.

Thanks!

Can you please release the download link for the raw protein data? I would appreciate it if I could use your code to understand how the raw data is processed.

Dear Sir/Madam,

My update:
As I don't have the completed dataset, I guess the original issue comes from below reasons:

1. npy_file_new(human_dataset).npy has 22217 data
2. Current available human data is only 4444+1111=5555
   Above causes the below problem. Please feel free to correct me. Thanks.

Original issue:
I am running this project on google colab. This might not be an issue, but I don't know how to solve it.
There is a problem showing as : IndexError: list index out of range.
The part of result as:
GCNN Loaded Training on 4444 samples..... 15657 first prot is /content/gdrive/MyDrive/PPI_GNN/PPI_GNN/human_features/processed/3AIH.pt [] 15657 Second prot is /content/gdrive/MyDrive/PPI_GNN/PPI_GNN/human_features/processed/1DEV.pt Traceback (most recent call last): File "train.py", line 97, in <module> train(model, device, trainloader, optimizer, epoch+1) File "train.py", line 45, in train for count,(prot_1, prot_2, label) in enumerate(trainloader): File "/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py", line 530, in __next__ data = self._next_data() File "/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py", line 570, in _next_data data = self._dataset_fetcher.fetch(index) # may raise StopIteration File "/usr/local/lib/python3.7/dist-packages/torch/utils/data/_utils/fetch.py", line 49, in fetch data = [self.dataset[idx] for idx in possibly_batched_index] File "/usr/local/lib/python3.7/dist-packages/torch/utils/data/_utils/fetch.py", line 49, in <listcomp> data = [self.dataset[idx] for idx in possibly_batched_index] File "/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataset.py", line 471, in __getitem__ return self.dataset[self.indices[idx]] File "/content/gdrive/MyDrive/PPI_GNN/PPI_GNN/data_prepare.py", line 41, in __getitem__ prot_1 = torch.load(glob.glob(prot_1)[0]) IndexError: list index out of range

The error comes from the code:
def __getitem__(self, index): prot_1 = os.path.join(self.processed_dir, self.protein_1[index]+".pt") print(index) print(f'first prot is {prot_1}') print(glob.glob('prot_1')) prot_2 = os.path.join(self.processed_dir, self.protein_2[index]+".pt") print(index) print(f'Second prot is {prot_2}') prot_1 = torch.load(glob.glob(prot_1)[0]) print(f'Here lies {glob.glob(prot_2)}') prot_2 = torch.load(glob.glob(prot_2)[0]) print(torch.tensor(self.label[index])) return prot_1, prot_2, torch.tensor(self.label[index])

It seems that glob.glob('prot_1') is null. How to solve this problem?
Thanks in advance.

Hi jha,

It is an excellent work for prediction PPI. I want to reproduce the prediction results and have a comparison in my work. However, the S. cerevisiae processed data in google drive is access denied. Can you share the data again?
Thanks very much!

Honchkrow