DL4DR: a flask-based web server to predict cancer drug response using deep learning
Frequently Asked Questions
DL4DR is a Flask framework-based web server that can be used to predict drug responses based on their chemical structures (represented by SMILES strings) and
cancer cell genomics profiles.
DL4DR employs cutting-edge deep neural network algorithms
for ultra-high throughput screening of most, if not all, cancer cell lines based on the big data of current measured drug responses and cancer cell line genomics. We expect that the
impact of such disruptive technology will be enormous as it significantly accelerates identification and rational design of novel potent therapeutics for either common or rare cancers,
by minimizing the need of experimental testing in drug discovery and development. It becomes especially powerful when used as the primary screen to pick a few cell lines and then
followed by HTS which can serve as the secondary screen to validate the predictions. This will advance projects forward with significantly high efficiency and low cost.
There are two ways to provide DL4DR with the input data as described below, and a more detailed instruction is linked to the "Help" button above in the submission form.
- In the online submission form box of the server, directly input the compound information in the format of compound name and SMILES separated by space, on each line. Examples are provided in the submission form box. Users also have the option to including or not to including cellular genomics information for predictions.
- Upload a file containing all compounds in the format as described above. The file should have the extension ".csv". An example file is also provided.
The input should be pretty self-explanatory and straightforward. An error message will show up in the submission form box if something is wrong. Please see
Help for more details.
DL4DR is free for academic research with a user-friendly interface. Currently, no registration is needed to explore its full functionality. However, commercial users should
contact us for agreement and support.
If you used DL4DR for your work, please cite this web page, or our manuscript once it is published.
Please contact us at info[at]imdlab[dot]org, or fill out this form.