Elucidating the Utility of Genomic Elements with Neuralnets¶
EUGENe is a Python toolkit for building and evaluating sequence-based deep learning models in genomics. It provides a unified workflow for managing data, training models, and interpreting predictions on biological sequences. Here are some ways you can use EUGENe:
Learn the fundamentals through practice. Replicate published studies and gain a deep understanding of deep learning models using our streamlined workflow.
Torture existing models. Discover the limitations and novel behaviors of existing models on new or synthetic datasets.
Apply existing architectures to new data. Not sure how to build a model for your dataset? Use EUGENe to try out established architectures.
Build new architecturess. Compare your custom models against existing architectures on known datasets.
EUGENe is designed to be used via its Python API, ideally within a notebook interface like Jupyter.
Getting started¶
(Optional) Read through the usage principles to get a better understanding of how EUGENe works in practice
Check out the basic usage tutorial for an example of how to run an end-to-end EUGENe workflow
Browse the main API page to see all the functionality that EUGENe provides
Note
EUGENe is a dynamic project and we’re constantly improving and adding new features based onuser feedback. If you encounter any issues or have suggestions for enhancements, we encourage you to open an issue on the EUGENe GitHub. Your feedback helps us make EUGENe better for everyone!
If you use EUGENe for your research, please cite our preprint: Klie et al. bioRxiv 2022
Contributing¶
EUGENe is an open-source project and we welcome contributions from the community. If you are interested in contributing, please see the contributor’s guide.