User's Manual

The flow diagram of using our server

Now we upload the sample test file to show how to use our server.

1. Login

For registered user, you can sign in with your username and password. For testers, click on the "Instant entry" button. If you forget your password, you can reset your password with your mailbox by clicking on 'Forgot Password?'.

We use a system of user management in order to give every user a separate space to handle their uploaded drugs. Users can register freely and easily by clicking on the "Register" button. If you do not want to register, just click the "Instant entry" button, You'll automatically login as the 'guest' user.

2. Upload a drug molecule

After logging in, you will see a page which allows you to upload a drug molecular.

Click on the "Browse" button to upload a drug file. We support multiple file formats including mol, mol2, pdb, pdbqt and sdf (How to generate and prepare a molecular file?). Usually, our server could check the suitability of your molecule. If the file suitability does not meet our demand, the results would be incredible. In order to protect the accessibility of other users, your uploaded file should not exceed the limitation of 50 kb. You can download our sample file naphthylamine.mol2 and upload it for a quick test.

3. Wait the interactome of your drug molecule to be constructed (several hours)

When your drug molecule file is uploaded correctly, our program starts to put your molecule in a queue automatically. Our server implements Autodock Vina3 to generate the docking scores of your molecule towards our library targets. You can click on the "Refresh" link to see the progress. If you upload our example file, it takes only several minutes to complete, however, for others it might take much longer.

If you are a registered user, you will see a list of your submissions like below. Click on the "Detail" button, you will first see detailed information of this drug.

If you login as the guest user (by clicking on the QuickLogin botton), you will be inform a special link to your access submission. Please bookmark this link for later access. If you input your email address before submission, the notification email would also include this special link.

4. Check the predicted indications of your molecule

After the docking progress is finished, you can see the predicted indications of your drug molecule along with their ICD9 codes. You will not see the predicted indications or the interactome of your molecule across the targetable proteins unless the progress rate reaches 100%.

In this page, you can see a list of predicted indications as disease names (with ICD9 codes in the front) as well as the estimated confidences in the parentheses (a value from 0 to 1) based on the interaction profile of your molecule. The diseases are grouped by their ICD9 families. You can enter a keyword in the "Search" box and press enter, to search for a disease or its ICD9 code. You can click on the "Reset" button to cancel your searching operation. The indication predictions of the user molecule are generated by machine learning models trained with 1,200+ FDA-approved drugs. The server suggests potential indications with estimated confidences across several hundred different disease indications. You can also download the whole table by simply clicking on the "download" link, and you can drag the downloaded text file into Excel to view the table.

In the right column of this page, you can view the basic information and 3D model of your molecular. By clicking on the button "See Chemical-Protein Interactome" at the right column, you can view candidate off-targets that potentially interact with your molecule.

5. Check the candidate off-targets tend to interact with your molecule

In step 4, Click on the "See Chemical-Protein Interactome" button to see candidate off-targets that potentially interact with your molecule. The server suggests candidate off-targets that potentially interact with your molecule. You can click on the "Visualization" button to visualize the binding pattern, with amino acid residues around 6.4Å of the drug molecule colored in the Applet.

The Docking Score fields represent the interaction strength of the molecule to the protein.

We are now using JSMol as a HLML5 molecular visualizer without requiring Java platform. It is capable on new versions of Firefox, Chrome and Internet Explorer 9+. For Internet Explorer 8, it is an out of date browser and the visualizer will be Java-based JMol. We recommend you either upgrade your browser or install Java Runtime Environment (JRE). You can choose "label on/off" to show/hide the labels of the amino acid residues around 6.4Å of the drug molecule.

6. Log out

If you want to logout or switch to another account, you can click "Log out" in the left navigation menu, or close your browser, the session will automatically end.


1 L. Yang et al, SePreSA: a server for the prediction of populations susceptible to serious adverse drug reactions implementing the methodology of a chemical-protein interactome. Nucleic Acids Research (Web Server issue). 2009;37:W406-12.

2 H. Luo et al, DRAR-CPI: a server for identifying drug repositioning potential and adverse drug reactions via the chemical-protein interactome. Nucleic Acids Research (Web Server issue). 2011;1(7).

3 H. Luo et al, DDI-CPI, a server that predicts drug-drug interactions through implementing the chemical-protein interactome. Nucleic Acids Research (Web Server issue). (2014): gku433.

4 O. Trott, A. J. Olson, AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization and multithreading. Journal of Computational Chemistry. 2010;31:455-46.