### Setup MixTCRpred Environment in Google Colab

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/colab_MixTCRpred.ipynb

This snippet prepares the Google Colab environment by cloning the MixTCRpred repository, installing necessary Python packages (wget, pytorch_lightning), and navigating into the project directory. This setup is required to run the notebook's examples.

```python
#Run this cell if using Google COLAB

#clone the repository and the data to run the notebook on Google Colab
!git clone https://github.com/GfellerLab/MixTCRpred
!pip install wget >> /dev/null 
!pip install pytorch_lightning >> /dev/null 
%cd MixTCRpred
```

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

### Test MixTCRpred Installation with Help Command

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/README.md

Runs the MixTCRpred script with the '--help' flag to display usage instructions and confirm successful installation.

```Bash
python MixTCRpred.py --help
```

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

### Install Python Dependencies for MixTCRpred

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/README.md

Upgrades pip and installs all required Python packages listed in 'requirements.txt' for MixTCRpred.

```Bash
pip install --upgrade pip
pip install -r requirements.txt
```

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

### Load Example Input TCRs

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/colab_MixTCRpred.ipynb

This Python code loads the 'test.csv' file, which contains example paired TCRs in the required input format for MixTCRpred, into a pandas DataFrame. This serves as a sample dataset for running predictions.

```python
input_tcrs = pd.read_csv("./test/test.csv")
input_tcrs
```

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

### MixTCRpred Command Line Arguments API

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/README.md

Documents the required and optional command-line arguments for the MixTCRpred script, including their purpose, format, and examples.

```APIDOC
Arguments:
  --model, -m [MixTCRpred_model_name]
    Description: The format is HLA_PeptideSequence (e.g. A0201_GILGFVFTL).
  --input, -i [input_TCR_file]
    Description: CSV file listing all the TCRs to test. See ./test/test.cvs for a reference input file. The columns order is not important. CDR3 alpha and beta should not be longer than 20 amino acids. Incomplete TCR entries are accepted, but the models will have lower predictive performance.
  --output, -o [output_file]
    Description: The name of the output file. It contains two extra columns than the input file: the MixTCRpred binding score and the %rank.

Optional Arguments:
  --list_models
    Description: To list the 146 MixTCRpred models for which we can currently run predictions. Models with less than 50 training TCRs have low confidence.
  --batch_size
    Description: The default batch size for testing is 1. If you have a large dataset of TCRs to test, increasing the batch_size can speed MixTCRpred up.
  --download model_name
    Description: To download a specific pretrained MixTCRpred model.
  --download_all
    Description: To download the 146 pretrained MixTCRpred models.
  --download_high
    Description: To download the 43 high-confidence pretrained MixTCRpred models.
```

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

### Download All MixTCRpred Pretrained Models

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/README.md

Downloads all 146 available pretrained MixTCRpred models.

```Bash
python MixTCRpred.py --download_all
```

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

### Basic MixTCRpred Command Line Usage

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/README.md

Shows the fundamental command structure for running MixTCRpred, requiring model name, input file, and output file.

```Bash
python MixTCRpred.py --model [MixTCRpred_model_name] --input [input_TCR_file] --output [output_file]
```

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

### Download Specific MixTCRpred Pretrained Model

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/README.md

Downloads a specified pretrained MixTCRpred model (e.g., A0201_NLVPMVATV) from the Zenodo dataset.

```Bash
python MixTCRpred.py --download A0201_NLVPMVATV
```

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

### Clone MixTCRpred GitHub Repository

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/README.md

Clones the MixTCRpred repository from GitHub to your local machine and navigates into the project directory.

```Bash
git clone https://github.com/GfellerLab/MixTCRpred 
cd MixTCRpred
```

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

### Create and Activate Python Virtual Environment

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/README.md

Creates a Python virtual environment named 'MixTCRpred_venv' and activates it for Unix/Mac OS users. This isolates project dependencies.

```Bash
python -m venv MixTCRpred_venv  
source MixTCRpred_venv/bin/activate
```

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

### Download High-Confidence MixTCRpred Models

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/README.md

Downloads the 43 high-confidence pretrained MixTCRpred models (those with more than 50 training TCRs).

```Bash
python MixTCRpred.py --download_high
```

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

### List Available MixTCRpred Models

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/README.md

Executes the MixTCRpred script with '--list_models' to display all currently available prediction models.

```Bash
python MixTCRpred.py --list_models
```

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

### Run MixTCRpred with Downloaded Model

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/colab_MixTCRpred.ipynb

This command-line snippet executes MixTCRpred using a newly downloaded pre-trained model (e.g., for HLA-A*02:01, YLQPRTFLL) to predict TCR binding for the input TCRs and store results in a new output file.

```python
!python MixTCRpred.py --model A0201_YLQPRTFLL --input ./test/test.csv --output ./test/out_A0201_YLQPRTFLL.csv
```

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

### Download MixTCRpred Pre-trained Model

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/colab_MixTCRpred.ipynb

This command-line snippet downloads a specific MixTCRpred pre-trained model (e.g., for HLA-A*02:01, YLQPRTFLL) that is not stored on GitHub. This step is necessary before making predictions with a new pMHC model.

```python
!python MixTCRpred.py --download A0201_YLQPRTFLL
```

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

### Load MixTCRpred Pre-trained Models Information

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/colab_MixTCRpred.ipynb

This Python code loads the 'info_models.csv' file into a pandas DataFrame. This file contains essential information about the available MixTCRpred pre-trained models, which can be used for various pMHC targets.

```python
import pandas as pd
import wget
import os
df_info = pd.read_csv("./pretrained_models/info_models.csv")
df_info
```

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

### Run MixTCRpred for TCR Prediction

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/README.md

Demonstrates how to run MixTCRpred to predict TCRs targeting a specific epitope (HLA-A*02:01, GILGFVFTL) using an input CSV file and saving results to an output CSV.

```Bash
python MixTCRpred.py --model A0201_GILGFVFTL --input ./test/test.csv --output ./test/output.csv
```

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

### Run MixTCRpred for TCR-pMHC Prediction

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/colab_MixTCRpred.ipynb

This command-line snippet executes the MixTCRpred script to predict TCR binding for a specific pMHC (e.g., HLA-A*02:01 GILGFVFTL). It requires a model name, an input TCR list, and an output file path. The leading '!' allows execution in a notebook environment.

```python
!python MixTCRpred.py --model A0201_GILGFVFTL --input ./test/test.csv --output ./test/out_A0201_GILGFVFTL.csv
```

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

### Create Permanent Alias for MixTCRpred

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/README.md

Creates a shell alias for MixTCRpred, making it executable from any directory using the Python version from the virtual environment. This alias can be made permanent by adding it to your .bashrc file.

```Bash
alias MixTCRpred='/home/[...]/MixTCRpred/MixTCRpred_venv/bin/python /home/[...]/MixTCRpred/MixTCRpred.py'
```

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

### Configure Absolute Path for Pretrained Models

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/README.md

Modifies the 'path_pretrained_models' variable in 'MixTCRpred.py' to use an absolute path, allowing the script to be run from any directory.

```Python
#change
path_pretrained_models = './pretrained_models'
#to 
path_pretrained_models = '/home/[...]/MixTCRpred/pretrained_models'
```

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

### Load Second MixTCRpred Prediction Results

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/colab_MixTCRpred.ipynb

This Python code loads the prediction results from the second specified output CSV file (e.g., './test/out_A0201_YLQPRTFLL.csv') into a pandas DataFrame for analysis, similar to the previous prediction results.

```python
results_LLW = pd.read_csv("./test/out_A0201_YLQPRTFLL.csv", comment = '#')
results_LLW
```

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

### Load MixTCRpred Prediction Results

Source: https://github.com/gfellerlab/mixtcrpred/blob/main/colab_MixTCRpred.ipynb

This Python code loads the prediction results from the specified output CSV file (e.g., './test/out_A0201_GILGFVFTL.csv') into a pandas DataFrame for analysis. The results are sorted from the most probable binder to the least probable binder.

```python
results_GIL = pd.read_csv("./test/out_A0201_GILGFVFTL.csv", comment = '#')
results_GIL
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.