### Install and Initialize Graphein

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/protein_tensors.ipynb

Installs the Graphein library and sets the verbosity level to False. This is a common setup step for using Graphein.

```python
#!pip install graphein
import graphein
graphein.verbose(False)
```

--------------------------------

### Install Graphein

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/alphafold_protein_graph_tutorial.ipynb

Install the graphein library if it is not already installed. This is a prerequisite for running the subsequent code examples.

```python
# Install graphein if necessary
#!pip install graphein
```

--------------------------------

### Setup FoldCompLightningDataModule

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/foldcomp.ipynb

Prepare the datasets by calling the setup method on the initialized data module.

```python
data_module.setup()
```

--------------------------------

### Install Requirements

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/tdc_developability.ipynb

Installs necessary libraries for Graphein and PyTDC. Run this if you haven't already.

```python
# Install requirements if necessary
# !pip install graphein
# !pip install PyTDC
```

--------------------------------

### Install Graphein and DSSP

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/protein_graph_analytics.ipynb

Install the Graphein library and the DSSP tool, which is necessary for computing solvent accessibility features. Use the provided commands for Colab or local installations.

```python
# Install graphein if necessary:
# !pip install graphein
# Install DSSP if necessary:
# !sudo apt-get install dssp (better for colab) OR !conda install -c salilab dssp
```

--------------------------------

### Install Graphein and PyMol

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/atom_graph_tutorial.ipynb

Installs Graphein and PyMol. PyMol is used for initial visualization in this tutorial.

```python
# Install Graphein if necessary
# !pip install graphein

# Install pymol if necessary - in this tutorial PyMol is only used for the initial plot. Feel free to skip!
# sudo apt-get install pymol (recommended for colab) OR conda install -c schrodinger pymol
```

--------------------------------

### Install Graphein and DSSP

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/subgraphing_tutorial.ipynb

Install Graphein and DSSP if necessary. DSSP is recommended for Colab environments.

```python
# Install Graphein if necessary
# !pip install graphein
# Install DSSP if necessary
# !sudo apt-get install dssp (better for colab) OR !conda install -c salilab dssp
```

--------------------------------

### Initialize and Setup FoldComp DataModule

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/foldcomp.ipynb

Instantiate the FoldCompLightningDataModule with dataset parameters and then call `setup()` to prepare the data splits and transformations for training.

```python
data_module = FoldCompLightningDataModule(
    data_dir="./test_data/",
    database="afdb_swissprot_v4",
    batch_size=32,
    num_workers=4
    )

data_module.setup()
```

--------------------------------

### Install PyTorch3D

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/protein_mesh_tutorial.ipynb

Install PyTorch3D, a library for 3D computer vision that Graphein utilizes for mesh processing and visualization.

```bash
# Install pytorch3d
# pip install pytorch3d
```

--------------------------------

### Install Graphein

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/interactive_plotly_example.ipynb

Install Graphein using pip if it's not already installed. This is a prerequisite for using Graphein's functionalities.

```python
# Install Graphein if necessary
# !pip install graphein
```

--------------------------------

### Install PyTorch using Conda

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Install PyTorch using the PyTorch Conda channel.

```bash
conda install -c pytorch pytorch
```

--------------------------------

### Set up vmd-python Library

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Clones and installs the vmd-python library from its GitHub repository.

```bash
# Set up vmd-python library
git clone https://github.com/Eigenstate/vmd-python.git
cd vmd-python
python setup.py build
python setup.py install
cd ..
```

--------------------------------

### Test getcontacts Installation

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Runs a test command to verify the getcontacts installation by calculating dynamic contacts.

```bash
# Test installation
cd getcontacts/example/5xnd
get_dynamic_contacts.py --topology 5xnd_topology.pdb \
                        --trajectory 5xnd_trajectory.dcd \
                        --itypes hb \
                        --output 5xnd_hbonds.tsv
```

--------------------------------

### Install Graphein using Pip

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Use pip for the base installation or to include additional featurization dependencies.

```bash
pip install graphein         # For base install
pip install graphein[extras] # For additional featurisation dependencies
```

--------------------------------

### Install GetContacts and Add to PATH

Source: https://github.com/a-r-j/graphein/blob/master/README.md

Install GetContacts using Git clone and add its directory to your PATH for use as an alternative intramolecular interaction computation tool.

```bash
git clone https://github.com/getcontacts/getcontacts

# Add folder to PATH
echo "export PATH=$PATH:`pwd`/getcontacts" >> ~/.bashrc
source ~/.bashrc
```

--------------------------------

### Test GetContacts Installation

Source: https://github.com/a-r-j/graphein/blob/master/README.md

Run a test command to verify the GetContacts installation by computing hydrogen bonds for a sample protein structure.

```bash
cd getcontacts/example/5xnd
get_dynamic_contacts.py --topology 5xnd_topology.pdb \
                        --trajectory 5xnd_trajectory.dcd \
                        --itypes hb \
                        --output 5xnd_hbonds.tsv
```

--------------------------------

### Install PyMol

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/protein_mesh_tutorial.ipynb

Install PyMol, which is used by Graphein for initial mesh generation. This step is recommended for Colab environments.

```bash
# Install pymol if necessary - in this tutorial PyMol is only used for the initial plot. Feel free to skip!
# sudo apt-get install pymol (recommended for colab) OR conda install -c schrodinger pymol
```

--------------------------------

### Install Graphein using Conda

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Clone the repository and use the provided environment file for Conda installation.

```bash
git clone https://www.github.com/a-r-j/graphein
cd graphein
conda env create -f environment.yml
pip install .
```

--------------------------------

### Install Graphein and PyTorch Geometric

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/creating_datasets_from_the_pdb.ipynb

Installs the necessary libraries for using Graphein and PyTorch Geometric. Ensure you have a compatible environment.

```python
!pip install graphein
!pip install torch_geometric
```

--------------------------------

### Install Graphein and DSSP

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/higher_order_graphs.ipynb

Installs the Graphein library and the DSSP tool, which is necessary for secondary structure analysis. Recommended for Colab environments.

```python
# Install Graphein if necessary
# !pip install graphein

# Install DSSP if necessary
# !sudo apt-get install dssp (better for colab) OR !conda install -c salilab dssp
```

--------------------------------

### PPISP Graph Initialization Example 3

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a PPISP graph, detecting nodes and calculating edges. This example shows the structure for a protein with 522 nodes and 470 edges, followed by a calculation for 215 edges.

```python
DGLGraph(num_nodes=95, num_edges=195,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
Detected 522 total nodes
Contact file found: ['ppisp_contacts/4fq0_contacts.tsv']
Calculated 470 intramolecular interaction-based edges
DGLGraph(num_nodes=182, num_edges=470,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### PPISP Graph Initialization Example 1

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a PPISP graph with specific node and edge data schemes, detecting nodes and calculating edges. This example shows the structure for a protein with 601 nodes and 827 edges.

```python
ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
Detected 601 total nodes
Contact file found: ['ppisp_contacts/4f6u_contacts.tsv']
Calculated 827 intramolecular interaction-based edges
DGLGraph(num_nodes=267, num_edges=827,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### PPISP Graph Initialization Example 2

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a PPISP graph, detecting nodes and calculating edges. This example shows the structure for a protein with 713 nodes and 742 edges, followed by a calculation for 195 edges.

```python
Detected 713 total nodes
Contact file found: ['ppisp_contacts/4fou_contacts.tsv']
Calculated 742 intramolecular interaction-based edges
DGLGraph(num_nodes=262, num_edges=742,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
Detected 713 total nodes
Contact file found: ['ppisp_contacts/4fou_contacts.tsv']
Calculated 195 intramolecular interaction-based edges
```

--------------------------------

### Install Graphein and PyTDC

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/molecule_model_tutorial_tox.ipynb

Install the necessary libraries for Graphein and the Total Drug Combinatorial Library (TDC). Ensure PyTorch and related libraries are configured based on your system's requirements (e.g., CUDA).

```python
# Install Graphein if necessary
# !pip install graphein[extras]

# Install TDC if necessary
# !pip install PyTDC

# NB you may need to install DL libraries such as pytorch, pytorch-lightning and torch-geometric
# These are left to the user to configure as they depend on your particular desired configuration (e.g CUDA)
```

--------------------------------

### DGLGraph Initialization Example 5

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a DGLGraph for a protein, showing node and edge data schemes after detecting nodes and calculating edges. This example is for a protein with 82 nodes and 192 edges.

```python
Detected 174 total nodes
Contact file found: ['ppisp_contacts/2o2v_contacts.tsv']
Calculated 192 intramolecular interaction-based edges
DGLGraph(num_nodes=82, num_edges=192,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### Build and Run Graphein Docker Container (GPU)

Source: https://github.com/a-r-j/graphein/blob/master/README.md

Build and manage the Graphein Docker container for GPU usage. Ensure NVIDIA Container Toolkit is installed and install the package locally after entering the container.

```bash
docker-compose up -d --build # start the container
docker-compose down # stop the container
```

--------------------------------

### DGLGraph Initialization Example 2

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a DGLGraph for a protein, showing node and edge data schemes after detecting nodes and calculating edges. This example is for a protein with 158 nodes and 391 edges.

```python
Detected 315 total nodes
Contact file found: ['ppisp_contacts/2j3r_contacts.tsv']
Calculated 391 intramolecular interaction-based edges
DGLGraph(num_nodes=158, num_edges=391,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### Get Node Count from PDB (Fourth Example)

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/dataloader_tutorial.ipynb

Reads a PDB file, constructs a protein graph, and returns the total number of nodes. This example is for a fourth PDB file.

```python
graphein.protein.graphs.Graph.from_pdb(path="/home/arian/graphein/notebooks/2ll6.pdb").number_of_nodes()
```

--------------------------------

### Get Node Count from PDB (Third Example)

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/dataloader_tutorial.ipynb

Reads a PDB file, constructs a protein graph, and returns the total number of nodes. This example is for a third PDB file.

```python
graphein.protein.graphs.Graph.from_pdb(path="/home/arian/graphein/notebooks/1lds.pdb").number_of_nodes()
```

--------------------------------

### Get Node Count from PDB (Another Example)

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/dataloader_tutorial.ipynb

Reads a PDB file, constructs a protein graph, and returns the total number of nodes. This example demonstrates the process for a different PDB file.

```python
graphein.protein.graphs.Graph.from_pdb(path="/home/arian/graphein/notebooks/3eiy.pdb").number_of_nodes()
```

--------------------------------

### Set up getcontacts Library

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Clones the getcontacts library and adds its directory to the system's PATH.

```bash
# Set up getcontacts library
git clone https://github.com/getcontacts/getcontacts.git
echo "export PATH=`pwd`/getcontacts:$PATH" >> ~/.bash_profile
source ~/.bash_profile
```

--------------------------------

### Start Graphein GPU Docker Container

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Build and run the Graphein GPU Docker container in detached mode. Ensure NVIDIA Container Toolkit is installed for GPU usage.

```bash
docker-compose up -d --build        # start the container
```

--------------------------------

### Start Graphein CPU Docker Container

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Build and run the Graphein CPU Docker container in detached mode.

```bash
docker-compose.cpu up -d --build    # start the container
```

--------------------------------

### Loading PDB Structure and Initializing Graph

Source: https://github.com/a-r-j/graphein/blob/master/datasets/pscdb/example_model_nb.ipynb

This snippet demonstrates downloading a PDB structure, detecting nodes, and initializing a graph. It also shows the path to the PDB file.

```python
Downloading PDB structure '1edz'...
Downloaded PDB file for: 1edz
Detected 0 HETATOM nodes
Detected 317 total nodes
1edz
['../../examples/pdbs/1edz.pdb']
PDB file detected: ../../examples/pdbs/1edz.pdb
```

--------------------------------

### Install PyTorch 1.9.0

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/protein_mesh_tutorial.ipynb

Install the specific version of PyTorch (1.9.0) required for PyTorch3D compatibility. Ensure this version is installed before proceeding.

```bash
# Install torch 1.9.0
# pip install torch==1.9.0
```

--------------------------------

### Install Graphein and Dependencies

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/dataloader_tutorial.ipynb

Install Graphein, PyTorch, and PyTorch Geometric if they are not already installed. Ensure compatibility between PyTorch and PyTorch Geometric versions.

```python
# Install graphein if necessary
# !pip install graphein

# Install torch if necessary. See https://pytorch.org/get-started/locally/
# pip install torch==1.11.0

# Install torch geometric if necessary. See: https://pytorch-geometric.readthedocs.io/en/latest/notes/installation.html
# pip install torch-scatter torch-sparse torch-cluster torch-spline-conv torch-geometric -f https://data.pyg.org/whl/torch-1.11.0+cpu.html
```

--------------------------------

### Initialize PyTorch Lightning Trainer

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/tdc_developability.ipynb

Sets up the trainer with specific configurations for GPU usage, benchmarking, and epoch limits. Use this to control the training environment.

```python
#NBVAL_SKIP
trainer = pl.Trainer(
    strategy=None,
    gpus=1,
    benchmark=True,
    deterministic=False,
    num_sanity_val_steps=0,
    max_epochs=10,
)
```

--------------------------------

### Install vmd-python Library (Linux)

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Installs the vmd-python library using conda on Linux.

```bash
# Set up vmd-python library
conda install -c https://conda.anaconda.org/rbetz vmd-python
```

--------------------------------

### PDB Download and Graph Construction

Source: https://github.com/a-r-j/graphein/blob/master/datasets/pscdb/example_model_nb.ipynb

This example shows the process of downloading a PDB structure, detecting nodes, and then computing contacts to form a DGLGraph. It includes the file path to the PDB and the resulting graph details.

```python
Output:
Downloading PDB structure '1qim'...
Downloaded PDB file for: 1qim
Detected 0 HETATOM nodes
Detected 532 total nodes
1qim
['../../examples/pdbs/1qim.pdb']
PDB file detected: ../../examples/pdbs/1qim.pdb
```

```python
Output:
Computed Contacts for: 1qim
Calculated 1486 intramolecular interaction-based edges
DGLGraph(num_nodes=1064, num_edges=1486,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
Detected 0 HETATOM nodes
Detected 456 total nodes
```

--------------------------------

### Protein Graph Initialization with Contact Data (Example 2)

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a protein graph with different contact data. Displays node and edge schemes, detected nodes, contact file, and calculated intramolecular edges.

```python
Detected 406 total nodes
Contact file found: ['ppisp_contacts/3vu9_contacts.tsv']
Calculated 564 intramolecular interaction-based edges
DGLGraph(num_nodes=215, num_edges=564,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
Detected 406 total nodes
Contact file found: ['ppisp_contacts/3vu9_contacts.tsv']
Calculated 538 intramolecular interaction-based edges
DGLGraph(num_nodes=191, num_edges=538,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### Protein Graph Initialization with Contacts

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

This example demonstrates initializing a Protein Graph class with contact file information. It outputs detected nodes, contact file paths, calculated edges, and the resulting DGLGraph structure.

```python
Detected 414 total nodes
Contact file found: ['ppisp_contacts/3p8b_contacts.tsv']
Calculated 130 intramolecular interaction-based edges
DGLGraph(num_nodes=60, num_edges=130,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

```python
Detected 414 total nodes
Contact file found: ['ppisp_contacts/3p8b_contacts.tsv']
Calculated 325 intramolecular interaction-based edges
DGLGraph(num_nodes=147, num_edges=325,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

```python
Detected 226 total nodes
Contact file found: ['ppisp_contacts/3ph0_contacts.tsv']
Calculated 159 intramolecular interaction-based edges
```

--------------------------------

### Install vmd-python Dependencies (macOS)

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Installs vmd-python dependencies using conda and Homebrew on macOS.

```bash
# Install vmd-python dependencies
conda install netcdf4 numpy pandas seaborn  expat tk=8.5  # Alternatively use pip
brew install netcdf pyqt # Assumes https://brew.sh/ is installed
```

--------------------------------

### Install get_contact_ticc.py Dependencies

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Installs necessary Python packages for get_contact_ticc.py using conda and pip.

```bash
# Install get_contact_ticc.py dependencies
conda install scipy numpy scikit-learn matplotlib pandas cython seaborn
pip install ticc==0.1.4
```

--------------------------------

### Protein Graph Initialization with Final Contacts

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

This example shows the final stages of protein graph initialization, including detected nodes, contact files, calculated edges, and the resulting DGLGraph object.

```python
Detected 326 total nodes
Contact file found: ['ppisp_contacts/3pv6_contacts.tsv']
Calculated 256 intramolecular interaction-based edges
DGLGraph(num_nodes=112, num_edges=256,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

```python
Detected 286 total nodes
Contact file found: ['ppisp_contacts/3q87_contacts.tsv']
Calculated 293 intramolecular interaction-based edges
DGLGraph(num_nodes=122, num_edges=293,
```

--------------------------------

### Install get_contact_ticc.py Dependencies (Linux)

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Installs Python dependencies for get_contact_ticc.py using conda and pip on Linux.

```bash
# Install get_contact_ticc.py dependencies
conda install scipy numpy scikit-learn matplotlib pandas cython
pip install ticc==0.1.4
```

--------------------------------

### Protein Graph Initialization with Contact Data (Example 3)

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a protein graph with different contact data. Displays node and edge schemes, detected nodes, contact file, and calculated intramolecular edges.

```python
Detected 163 total nodes
Contact file found: ['ppisp_contacts/3vz9_contacts.tsv']
Calculated 266 intramolecular interaction-based edges
DGLGraph(num_nodes=103, num_edges=266,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
Detected 163 total nodes
Contact file found: ['ppisp_contacts/3vz9_contacts.tsv']
Calculated 118 intramolecular interaction-based edges
DGLGraph(num_nodes=60, num_edges=118,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### PPISP Graph Initialization Example 5

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Displays the structure of a PPISP graph with 71 nodes and 155 edges, including node and edge data schemes.

```python
DGLGraph(num_nodes=71, num_edges=155,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### Install PyTorch Geometric using Conda

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Install PyTorch Geometric using the specified Conda channels.

```bash
conda install pytorch-geometric -c rusty1s -c conda-forge
```

--------------------------------

### Protein Graph Initialization with Contact Data (Example 5)

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a protein graph with different contact data. Displays node and edge schemes, detected nodes, contact file, and calculated intramolecular edges.

```python
Detected 2008 total nodes
Contact file found: ['ppisp_contacts/3w0l_contacts.tsv']
Calculated 1742 intramolecular interaction-based edges
DGLGraph(num_nodes=598, num_edges=1742,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
```

--------------------------------

### Set up getcontacts Library (Linux)

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Clones the getcontacts library and updates the .bashrc file to include it in the PATH on Linux.

```bash
# Set up getcontacts library
git clone https://github.com/getcontacts/getcontacts.git
echo "export PATH=`pwd`/getcontacts:$PATH" >> ~/.bashrc
source ~/.bashrc
```

--------------------------------

### Install PyMol using Conda

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Install PyMol via Conda, necessary for visualizations and mesh generation.

```bash
conda install -c schrodinger pymol # Required for PyMol visualisations & mesh generation
```

--------------------------------

### Protein Graph Initialization with Contact Data (Example 4)

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a protein graph with different contact data. Displays node and edge schemes, detected nodes, contact file, and calculated intramolecular edges.

```python
Detected 2008 total nodes
Contact file found: ['ppisp_contacts/3w0l_contacts.tsv']
Calculated 1239 intramolecular interaction-based edges
DGLGraph(num_nodes=426, num_edges=1239,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### DGLGraph Initialization Example 5

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a DGLGraph with 115 nodes and 297 edges, detailing node and edge data schemes after detecting nodes and finding a contact file.

```text
DGLGraph(num_nodes=115, num_edges=297,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
Detected 645 total nodes
Contact file found: ['ppisp_contacts/3tu3_contacts.tsv']
Calculated 270 intramolecular interaction-based edges
```

--------------------------------

### Initialize FoldCompLightningDataModule

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/foldcomp.ipynb

Instantiate the data module with specified parameters for data directory, database, batch size, number of workers, and data splits.

```python
data_module = FoldCompLightningDataModule(
    data_dir="./test_data/",
    database="afdb_swissprot_v4",
    batch_size=32,
    num_workers=4,
    train_split=0.7,
    val_split=0.2,
    test_split=0.1
    )
```

--------------------------------

### Progress Bar Output Example 4

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Example of a progress bar output during a computation, indicating 67% completion.

```text
 67%|██████▋   | 282/422 [00:40<00:24,  5.83it/s]
```

--------------------------------

### Progress Bar Output Example 3

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Example of a progress bar output during a computation, indicating 66% completion.

```text
 66%|██████▋   | 280/422 [00:40<00:28,  4.90it/s]
```

--------------------------------

### Install Graphein Development Environment using Conda

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Clone the repository and create a Conda environment for development, including GPU builds.

```bash
git clone https://www.github.com/a-r-j/graphein
cd graphein
conda env create -f environment-dev.yml
pip install -e .  # Install in editable mode
```

--------------------------------

### Progress Bar Output Example 2

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Example of a progress bar output during a computation, indicating 66% completion.

```text
 66%|██████▌   | 278/422 [00:40<00:39,  3.66it/s]
```

--------------------------------

### Progress Bar Output Example 1

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Example of a progress bar output during a computation, indicating 66% completion.

```text
 66%|██████▌   | 277/422 [00:39<00:35,  4.04it/s]
```

--------------------------------

### DGLGraph Initialization Example 4

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a DGLGraph with 208 nodes and 432 edges, detailing node and edge data schemes after detecting nodes and finding a contact file.

```text
DGLGraph(num_nodes=208, num_edges=432,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
Detected 2571 total nodes
Contact file found: ['ppisp_contacts/3tgx_contacts.tsv']
Calculated 432 intramolecular interaction-based edges
```

--------------------------------

### Install PyTorch3D using Conda

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Install PyTorch3D using the PyTorch Conda channel. Note that this requires fvcore and iopath.

```bash
conda install -c pytorch3d pytorch3d #  NB requires fvcore and iopath
```

--------------------------------

### Protein Graph Initialization with Multiple Contact Files

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes protein graphs from multiple contact files, showing the resulting node and edge counts for each graph.

```python
DGLGraph(num_nodes=119, num_edges=293,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
Detected 186 total nodes
Contact file found: ['ppisp_contacts/1z3e_contacts.tsv']
Calculated 176 intramolecular interaction-based edges
DGLGraph(num_nodes=67, num_edges=176,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
Detected 254 total nodes
Contact file found: ['ppisp_contacts/1z5y_contacts.tsv']
Calculated 247 intramolecular interaction-based edges
DGLGraph(num_nodes=118, num_edges=247,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
Detected 254 total nodes
Contact file found: ['ppisp_contacts/1z5y_contacts.tsv']
```

--------------------------------

### DGLGraph Initialization Example 7

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a DGLGraph with 526 nodes and 1510 edges, detailing node and edge data schemes after detecting nodes and finding a contact file.

```text
Detected 645 total nodes
Contact file found: ['ppisp_contacts/3tu3_contacts.tsv']
Calculated 1510 intramolecular interaction-based edges
DGLGraph(num_nodes=526, num_edges=1510,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### Install DSSP using Conda

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Install the DSSP package via Conda, which is required for computing secondary structural features.

```bash
conda install -c salilab dssp # Required for computing secondary structural features
```

--------------------------------

### DGLGraph Initialization Example 1

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a DGLGraph with 197 nodes and 447 edges, detailing node and edge data schemes.

```text
Calculated 447 intramolecular interaction-based edges
DGLGraph(num_nodes=197, num_edges=447,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### Install Graphein Development Dependencies using Pip

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Use pip to install Graphein with development dependencies or all available dependencies.

```bash
pip install graphein[dev] # For dev dependencies
pip install graphein[all] # To get the lot
```

--------------------------------

### DGLGraph Initialization Example 2

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a DGLGraph with 198 nodes and 592 edges, including node and edge data schemes.

```text
Calculated 592 intramolecular interaction-based edges
DGLGraph(num_nodes=198, num_edges=592,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### Graph Construction with HETATOM Nodes and Interaction Edges

Source: https://github.com/a-r-j/graphein/blob/master/datasets/pscdb/example_model_nb.ipynb

This example shows graph construction for a protein structure with a specific number of HETATOM nodes and total nodes, including the calculation of intramolecular interaction-based edges.

```python
Detected 21 HETATOM nodes
Detected 1345 total nodes
Contact file found: ['../../examples/contacts/2p0m_contacts.tsv']
Calculated 1850 intramolecular interaction-based edges
DGLGraph(num_nodes=663, num_edges=1850,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### Install DGL using Conda

Source: https://github.com/a-r-j/graphein/blob/master/docs/source/getting_started/installation.rst

Install the Deep Graph Library (DGL) using the DGL team's Conda channel.

```bash
conda install -c dglteam dgl
```

--------------------------------

### DGLGraph Initialization with Contact File Information

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Illustrates the initialization of a DGLGraph, including detected node counts and contact file information. This helps in understanding the data source and graph size.

```python
Detected 343 total nodes
Contact file found: ['ppisp_contacts/3r07_contacts.tsv']
Calculated 203 intramolecular interaction-based edges
DGLGraph(num_nodes=88, num_edges=203,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### Protein Graph Initialization and Edge Calculation (Variant 2)

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Demonstrates protein graph initialization with a different contact file and edge calculation. This output shows the DGLGraph structure after processing another protein.

```text
Detected 530 total nodes
Contact file found: ['ppisp_contacts/1k74_contacts.tsv']
Calculated 697 intramolecular interaction-based edges
DGLGraph(num_nodes=242, num_edges=697,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### Structural Interactomics Setup

Source: https://github.com/a-r-j/graphein/blob/master/notebooks/ppi_graph.ipynb

Imports and configuration for constructing and visualizing structural interactomes. This setup is necessary for analyzing protein-protein interactions at a structural level.

```python
import matplotlib.pyplot as plt
logging.getLogger("matplotlib").setLevel(logging.WARNING)

from graphein.protein.config import ProteinGraphConfig
from graphein.protein.graphs import construct_graph
from graphein.protein.utils import download_alphafold_structure
from graphein.protein.visualisation import plot_protein_structure_graph

pg_config = ProteinGraphConfig()
```

--------------------------------

### Protein Graph Initialization for 1e4k (First Instance)

Source: https://github.com/a-r-j/graphein/blob/master/datasets/ppisp/ppisp_example.ipynb

Initializes a protein graph for '1e4k', calculating edges from its contact file. Displays the DGLGraph structure.

```python
Detected 604 total nodes
Contact file found: ['ppisp_contacts/1e4k_contacts.tsv']
Calculated 882 intramolecular interaction-based edges
DGLGraph(num_nodes=432, num_edges=882,
         ndata_schemes={'id': Scheme(shape=(), dtype=dtype('O')), 'residue_name': Scheme(shape=(), dtype=dtype('O')), 'h': Scheme(shape=(7,), dtype=torch.float32), 'coords': Scheme(shape=(3,), dtype=torch.float32), 'ss': Scheme(shape=(9,), dtype=torch.float32), 'asa': Scheme(shape=(1,), dtype=torch.float32), 'rsa': Scheme(shape=(1,), dtype=torch.float32)}
         edata_schemes={'rel_type': Scheme(shape=(17,), dtype=torch.float64), 'norm': Scheme(shape=(), dtype=torch.float32)})
```

--------------------------------

### Install Optional Conda Dependencies

Source: https://github.com/a-r-j/graphein/blob/master/README.md

Install DSSP and PyMol using Conda, which are required for specific features like secondary structure computation and PyMol visualizations.

```bash
conda install -c salilab dssp # Required for computing secondary structural features
conda install -c schrodinger pymol # Required for PyMol visualisations & mesh generation
```