While a few UPS substrate mutations can be implicated in cancer based on known degrons, systematic investigation requires better degron annotation. To address this challenge, we developed a protein sequence-based model, deepDegron, that leverages data from the recently published high throughput global protein stability (GPS, (Koren et al., 2018; Timms et al., 2019)) assay of the n-terminal and c-terminal proteome to predict degrons. GPS measures the protein stability impact of peptides when attached to proteins (Yen et al., 2008), as measured by FACS-sorting of cells based on a fluorescent reporter protein (GFP, green) compared to a control reporter with no peptide attached (dsRed, red). Because the peptides consisted of known sequences and could contain degrons, deepDegron can learn sequence-rules of degron impact on protein stability.

For more documentation, please visit our website documentation.

We recommend that you use python 3.7 to run deepDegron.

The easiest way to install deepDegron is through pip.

$ pip install deepDegron
$ pyensembl install --release 75 --species human  # download human hg19 reference data
$ pyensembl install --release 95 --species human  # download human hg38 reference data

As a first step, please change to the top-level directory in the deepDegron source code.

You can install the dependencies of deepDegron using conda. Once you have conda installed, create an environment to run deep degron using the following commands:

$ conda env create -f environment.yml  # install dependencies
$ source activate deepDegron  # activate environment
$ pyensembl install --release 75 --species human  # download human reference data
$ python setup.py install  # install deepDegron

An alternative way to install the python dependencies is to use pip.

$ python -m pip install --upgrade pip
$ pip install -r requirements.txt  # install required packages
$ pyensembl install --release 75 --species human  # download human reference data
$ python setup.py install  # install deepDegron