Skip to content

Latest commit

 

History

History
38 lines (31 loc) · 903 Bytes

README.md

File metadata and controls

38 lines (31 loc) · 903 Bytes

DeepOM

DeepOM is a Python deep-learning software for optical mapping of DNA images to a genome reference. See the paper for details.

For inference only: Running the example script

Ensure you use python >= 3.10

cd tool
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
python example.py

Reproducing the figures from the paper

For training: Installation

conda env create -f environment.yml
conda activate deepom
pip install -r requirements.txt
pip install -e .
  • Reproducing the figures from the paper: See Jupyter notebooks in the figures dir.
  • Running the localizer net training:
    conda activate deepom
    python deepom/localizer.py
  • Running the benchmark:
    conda activate deepom
    python deepom/bionano_compare.py