DeepFrag is a deep-learning model that recommends strategies for lead optimization. The original implementation is a Python script and so requires a certain degree of computational expertise. To encourage broader adoption, we have created the DeepFrag browser app, which provides a user-friendly graphical user interface that runs the DeepFrag model in users' web browsers. The browser app does not require users to upload their molecular structures to a third-party server, nor does it require the separate installation of any third-party software. It can be accessed free of charge, without requiring registration, at http://durrantlab.com/deepfrag.
We have tested the DeepFrag library on macOS 10.14.5, Windows 10.0.19041 Home, and Ubuntu Linux 18.04.5 LTS.
|Chrome 88.0.4324.87||macOS 10.14.5|
|Firefox 84.0||macOS 10.14.5|
|Safari 13.1.1||macOS 10.14.5|
|Chrome 87.0.4280.141||Windows 10.0.19041 Home|
|Firefox 84.0.2||Windows 10.0.19041 Home|
|Edge 87.0.664.75||Windows 10.0.19041 Home|
|Chrome 87.0.4280.141||Android 10|
|Firefox 84.1.4||Android 10|
|Safari 14||iPhone SE iOS 14.3|
|Chromium 87.0.4280.141||Ubuntu Linux 18.04.5 LTS|
|Firefox 84.0.2||Ubuntu Linux 18.04.5 LTS|
src/: The DeepFrag source files. You cannot use these files directly. They must be compiled. For a compiled, ready-to-use copy of the DeepFrag browser app, see the Releases page.
tsconfig.json: Files used to compile the contents of the
src/directory to the
README.md: Documentation files.
After visiting the DeepFrag browser-app URL, users will first encounter the "Input Parameters" tab. In the "Input Receptor and Ligand Files" subsection, users can specify the protein receptor and ligand file for optimization in any of several popular molecular-model formats. The contents of these files are loaded into the browser's memory, but they are never transmitted/uploaded to any third-party server. Users who wish to simply test DeepFrag can instead click the "Use Example Files" button to load in a pre-prepared structure of Pin1p bound to a small-molecule ligand (PDB 2XP9).
The "Molecular Viewer" subsection contains a 3Dmol.js molecular viewer were the specified files are displayed. This subsection also includes two toggle buttons. The "Delete Atom" button allows users to remove ligand atoms from the structure by clicking on them. The "Select Atom as Growing Point" toggle button allows users to indicate which ligand atom should serve as the growing point (i.e., connection point) that connects the predicted fragments to the parent ligand molecule. After users click the appropriate ligand atom, a yellow transparent sphere indicates the location of the growing point.
At the bottom of the Input Parameters tab, the "Temporary Save" button saves the specified parameters (i.e., receptor/ligand files, connection point, etc.) to the browser's session storage. These same parameters can be later restored using the "Load Saved Data" button. Otherwise, the user simply clicks the "Start DeepFrag" button to begin the DeepFrag run. The browser implementation is identical to the production model, but without grid rotation/fragment averaging.
DeepFrag displays the "Output" tab once the calculations are complete. The "Visualization" subsection again displays the specified receptor, ligand, and growing point for user convenience. Below the molecular visualization, a table shows the SMILES strings, molecular structures, and DeepFrag scores of the top twenty predicted fragments, sorted from most to least promising.
The "Output Files" subsection allows users to directly view DeepFrag output text files. They can also press the associated "Download" buttons to save the files to disk. These files include a more complete list of the predicted fragments (TSV format), the 3D coordinates of the selected growing point (JSON format), and the receptor and ligand files used for analysis (PDB format).
The "Start Over" tab displays a simple button that allows the user to restart the DeepFrag app. A warning message reminds the user that they will lose the results of the current DeepFrag run unless they have saved their output files.
Most users will wish to simply access the already compiled, publicly available DeepFrag browser app. If you wish to instead run DeepFrag browser app on your own UNIX-like computer (LINUX, macOS, etc.), follow these instructions:
- Download the compiled app from the Releases page
- Uncompress the file:
- Change to the new
- Start a local server.
- You can use
npm install -g http-server
- Or Python:
- Python 3:
python3 -m http.server 8000
- Python 2 (out of date):
python -m SimpleHTTPServer 8000
- Python 3:
- You can use
- Access the server from your web-browser (e.g.,
Running DeepFrag on other operating systems (e.g., Windows) should be similar.
The vast majority of users will not need to compile the DeepFrag browser app on their own. Simply use the online version or download the already compiled files from the Releases page. If you need to make modifications to the source code, these instructions should help with re-compiling on UNIX-like systems:
- Clone or download the git repository:
git clone https://git.durrantlab.pitt.edu/jdurrant/deepfrag-app.git
- Change into the new
- Install the required
- Fix any vulnerabilities:
npm audit fix
- Make sure Python is installed system wide, and that
pythonworks from the command line (tested using Python 2.7.15)
- To deploy a dev server:
npm run start
- To compile the contents of
npm run build
In some circumstances, the DeepFrag browser app may report usage statistics to Google Analytics. These reports are useful for securing and justifying funding for the Durrant lab. Usage statistics are only sent if the browser-app URL contains the substring "durrantlab," so installing DeepFrag on your own server should prevent reporting. Even when using the publicly available version of DeepFrag hosted at http://durrantlab.com/deepfrag, information about your specific receptor and ligand files is never transmitted to any remote server.