Run_Colab.ipynb

{"cells":[{"cell_type":"markdown","metadata":{"id":"fwofdfUiT-cJ"},"source":["# <center> DeepLigType: Predicting Ligand Types of Protein-Ligand Binding Sites Using a Deep Learning Model </center>\n"]},{"cell_type":"markdown","metadata":{},"source":["#### * Resolving Fpocket Installation Issues with the Latest Colab Update (December 2023) \n","\n","The Colab runtime has been upgraded to CUDA 12.2 in December 2023. You can find the official Colab announcement here: [Link](https://github.com/googlecolab/colabtools/issues/4214#issue-2037844472) \n","\n","Currently, there is an issue with the installation process using Conda in Google Colab.\n","\n","Since Fpocket is also installed with conda, Fpocket cannot be used in the Google Colab version at the moment.\n","\n","Solution: Use the fallback runtime version of Google Colab.\n","\n","Tools-> Command Palette -> Use Fallback Runtime Version"]},{"cell_type":"markdown","metadata":{"id":"TO7SVE7bkEbj"},"source":["## * Install Libraries\n"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"AIJ-2y4z9_zA"},"outputs":[],"source":["# May need to be run twice. This is for fpocket installation\n","!pip install -q condacolab\n","import condacolab\n","condacolab.install()"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"U5spbxcL-Cm1"},"outputs":[],"source":["import condacolab\n","condacolab.check()"]},{"cell_type":"code","execution_count":null,"metadata":{},"outputs":[],"source":["import torch\n","\n","# Check if a GPU is available\n","if torch.cuda.is_available():\n","    # Get the name of the GPU\n","    gpu_name = torch.cuda.get_device_name(0)\n","    print(f\"GPU: {gpu_name}\")\n","else:\n","    print(\"No GPU available - You can't run the program\")"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"IsfKrX0KjlhB"},"outputs":[],"source":["!git clone https://github.com/drorhunvural/DeepLigType.git "]},{"cell_type":"code","execution_count":null,"metadata":{},"outputs":[],"source":["#To clone the project into your drive, please execute this cell.\n","\n","# from google.colab import drive\n","# drive.mount('/content/drive')\n","\n","# !git clone https://github.com/drorhunvural/DeepLigType.git '/content/drive/My Drive/your-target-directory'"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"dnmdfdxaUU8F"},"outputs":[],"source":["#it takes 5-6 mins.\n","!pip3 install torch torchvision torchaudio\n","!pip install molgrid\n","!pip install biopython\n","!pip install scikit-learn\n","!mamba install fpocket"]},{"cell_type":"code","execution_count":null,"metadata":{},"outputs":[],"source":["!pip install numpy==1.21.6 #Otherwise molgrid library is giving an error "]},{"cell_type":"markdown","metadata":{"id":"X4qZWFMcUh5P"},"source":["## * Predict.py"]},{"cell_type":"code","execution_count":null,"metadata":{},"outputs":[],"source":["#Note: The input PDB file must be included within the DeepLigType project !! e.g. .../DeepLigType/5kw2.pdb"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"mYTqZuyuUlTS"},"outputs":[],"source":["# Example: !python '/content/DeepLigType/predict.py' -p '/content/DeepLigType/5xpp.pdb' -t \"CNN_2023-08-30_acc_0.717969_59.26.pth\"\n","# Example: !python '/content/DeepLigType/predict.py' -p '/content/DeepLigType/5xpp.pdb' -t \"CBAM_2023-08-29_acc_0.915781_74.16.pth\"\n","# Example: !python '/content/DeepLigType/predict.py' -p '/content/DeepLigType/5xpp.pdb' -t \"DenseNet_2023-08-20_acc_0.930625_70.73.pth\"\n","# Example: !python '/content/DeepLigType/predict.py' -p '/content/DeepLigType/5xpp.pdb' -t \"SeResNet_2023-08-30_val_acc_0.881406_68.16.pth\"\n","# Example: !python '/content/DeepLigType/predict.py' -p '/content/DeepLigType/5xpp.pdb' -t \"ResNet18_2023-10-06_train_acc_0.883281.pth\"\n","\n","!python '/content/DeepLigType/predict.py' -p '/content/DeepLigType/xxxx.pdb' -t \"CBAM_2023-08-29_acc_0.915781_74.16.pth\""]},{"cell_type":"markdown","metadata":{"id":"nG9i2h9EkPwp"},"source":["## * Evaluation.py"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"cFbtfvqw1ngJ"},"outputs":[],"source":["#Note 1: Before running evaluation.py, please make sure you've downloaded 'pdb.molcache' (1.61 GB) and placed it in the 'dataset' folder within the project. \n","# You can find the download link in the provided README on GitHub."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"FlzoVpAP1tNV"},"outputs":[],"source":["#Note 2: If you wish to evaluate using other models such as CNN, SeResnet, etc., download from the link provided in the GitHub README."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"ShqAsTlckRvP"},"outputs":[],"source":["# Example: !python '/content/DeepLigType/evaluation.py' -t \"CNN_2023-08-30_acc_0.717969_59.26.pth\"\n","# Example: !python '/content/DeepLigType/evaluation.py' -t \"CBAM_2023-08-29_acc_0.915781_74.16.pth\"\n","# Example: !python '/content/DeepLigType/evaluation.py' -t \"DenseNet_2023-08-20_acc_0.930625_70.73.pth\"\n","# Example: !python '/content/DeepLigType/evaluation.py' -t \"SeResNet_2023-08-30_val_acc_0.881406_68.16.pth\"\n","# Example: !python '/content/DeepLigType/evaluation.py' -t \"ResNet18_2023-10-06_train_acc_0.883281.pth\"\n","\n","!python '/content/DeepLigType/evaluation.py' -t \"trainedmodels\""]},{"cell_type":"markdown","metadata":{"id":"IX1oRZFOU5bd"},"source":["## * Training.py"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"aWqkyNKXUs3i"},"outputs":[],"source":["# Example: !python '/content/DeepLigType/train.py' -m \"cbam\"\n","# Example: !python '/content/DeepLigType/train.py' -m \"seresnet\"\n","# Example: !python '/content/DeepLigType/train.py' -m \"resnet18\"\n","# Example: !python '/content/DeepLigType/train.py' -m \"cnn\"\n","# Example: !python '/content/DeepLigType/train.py' -m \"densenet\"\n","\n","!python '/content/DeepLigType/train.py' -m \"cnn\""]}],"metadata":{"accelerator":"GPU","colab":{"gpuType":"V100","provenance":[]},"kernelspec":{"display_name":"Python 3","name":"python3"},"language_info":{"name":"python"}},"nbformat":4,"nbformat_minor":0}