sizheqiu/dltkcat

Deep learning based prediction of temperature dependent enzyme turnover rates

DLTKcat

DLTKcat v1.0: Deep learning based prediction of temperature dependent enzyme turnover rates

Dataset curation from SABIO-RK and BRENDA

The dataset curation process is in /code/GetData.ipynb.

How to use DLTKcat ?

1. Required inputs: substrate name, Uniprot ID of enzyme protein, temperature.
   
2. Get SMILES strings and enzyme protein sequences using convert_input(path, enz_col, sub_col ) in /code/feature_functions.py.
   
3. The input must be a csv file with columns of 'smiles', 'seq', 'Temp_K_norm', 'Inv_Temp_norm'.
   'Temp_K_norm' and 'Inv_Temp_norm' are normalized temperature and inverse temperature values.
   
4. Run prediction:
   

python predict.py --model_path [default = /data/performances/model_latentdim=40_outlayer=4_rmsetest=0.8854_rmsedev=0.908.pth]<br>
--param_dict_pkl [default = /data/hyparams/param_2.pkl] <br>
--input [input.csv] --output [output file name] <br>
--has_label [default = False]

5. Get attention weights of protein residues:
   

python get_attention.py --input [input.csv] --output [output file name]

Case studies

1. Mutants of Pyrococcus furiosus Ornithine Carbamoyltransferase via directed evolution (/data/PFOCT/,/code/CaseStudy_PFOCT.ipynb).
   Ref: https://doi.org/10.1128/jb.183.3.1101-1105.2001
2. Growth and metabolism of Lactococcus lactis and Streptococcus thermophilus at different temperatures(/data/GEMs, /code/GEMs.ipynb).
   Ref: https://doi.org/10.1038/srep14199, https://doi.org/10.1111/j.1365-2672.2004.02418.x

Dependencies

1. Pytorch: https://pytorch.org/
2. Scikit-learn: https://scikit-learn.org/
3. RDKit:https://www.rdkit.org/
4. BRENDApyrser: https://github.com/Robaina/BRENDApyrser
5. COBRApy: https://github.com/opencobra/cobrapy
6. Seaborn statistical data visualization:https://seaborn.pydata.org/index.html
7. Escher: https://github.com/zakandrewking/escher

Citation

DLTKcat: deep learning based prediction of temperature dependent enzyme turnover rates Sizhe Qiu, Simiao Zhao, Aidong Yang bioRxiv 2023.08.10.552798; doi: https://doi.org/10.1101/2023.08.10.552798

Issue

Users might encounter "Index out of range" error at amino_vector = self.embedding_layer_amino(amino).
The potential solution is +1 to n_atom, n_amino in model parameters, and train a new model.

dltkcat's Issues

Some SMILES from E.coli dataset can not be correctly processed.

Hello!
When I predict kcat of some E.coli reactions, it says my SMILES are out of the range of atoms like this.
For example, CC1(CC(=O)O)C2=Cc3[nH]c(c(CCC(=O)O)c3CC(=O)O)Cc3[nH]c(c(CC(=O)O)c3CCC(=O)O)C=C3N=C(C=C(N2)C1CCC(=O)O)C(C)(CC(=O)O)C3CCC(=O)O

Traceback (most recent call last):
  File "predict.py", line 82, in <module>
    pred = M( atoms_pad, atoms_mask, adjacencies_pad, amino_pad, amino_mask, batch_fps, inv_Temp, Temp )
  File "/torch/nn/modules/module.py", line 1194, in _call_impl
    return forward_call(*input, **kwargs)
  File "/DLTKcat/code/DLTKcat.py", line 161, in forward
    atoms_vector = self.comp_gat(atoms, adjacency)
  File "/DLTKcat/code/DLTKcat.py", line 102, in comp_gat
    atoms_vector = self.embedding_layer_atom(atoms)
  File "/torch/nn/modules/module.py", line 1194, in _call_impl
    return forward_call(*input, **kwargs)
  File "/torch/nn/modules/sparse.py", line 162, in forward
    self.norm_type, self.scale_grad_by_freq, self.sparse)
  File "/torch/nn/functional.py", line 2210, in embedding
    return torch.embedding(weight, input, padding_idx, scale_grad_by_freq, sparse)
IndexError: index out of range in self

I use CPU here. If use GPU there will be similar errors.
Could you please tell me what's the problem with my operation?
Looking forward to your reply.
Thank you!

code/run_train_test.py

，is there should be atom_dict = load_pickle('../data/dict/atom_dict.pkl')

brenda.txt

Hi, @SizheQiu

Thanks for your great work, I want to know where could I get the file brenda.txt mentioned in the file GetData.ipynb?