DeepLearningForIonMobility

Installation

Begin by downloading the repo. TODO: better instructions here

Create a virtual environment using the following command and replacing "environment_name" with whatever you would like your environment to be called.

# Linux and macOS
python -m venv environment_name  
# Windows
python -m venv venv environment_name

To activate the environment so that you can install and use packages run:

# Linux and macOS
source virtual_environment_name/bin/activate
# Windows
.\virtual_environment_name\Scripts\activate

Now you can install the required packages from the requirements.txt file provided.

pip3 install -r requirements.txt

You can exit the environment using the 'deactivate' command.

# Linux, macOS, and Windows
deactivate

Training models

Some default parameters are specified in configs-default.yaml. Other parameters will be specified by a config file that you create. You may also overwrite the default values in your own config file.

python run_training.py <path to training data> <path to testing data> <path to config file>

Making predictions using pre-trained models

###Formatting data Peptides are represented as strings of single letter amino acid codes. Supported modified amino acids include .... represented as .... respectively.

###Download the model

###Make predictions You can use the following command to make predictions.
python predict.py <path to data file>