Protein Structure and Function Prediction

RaptorX is a protein structure prediction server developed by Xu group, excelling at predicting 3D structures for protein sequences without close homologs in the Protein Data Bank (PDB). Given an input sequence, RaptorX predicts its secondary and tertiary structures as well as solvent accessibility and disordered regions. RaptorX also assigns the following confidence scores to indicate the quality of a predicted 3D model: P-value for the relative global quality, GDT (global distance test) and uGDT (un-normalized GDT) for the absolute global quality, and RMSD for the absolute local quality of each residue in the model. RaptorX-Binding is a web server that predicts the binding sites of a protein sequence, based upon the predicted 3D model by RaptorX. More details can be found HERE.

Currently, we have the following servers available.
RaptorX Structure Prediction: predict 3-state and 8-state secondary structure, solvent accessibility, disordered regions and tertiary structure for a protein sequence
[Submit] [Find jobs by: ID or sequence, Email] [Example] [Refs: 1, 2, 3, 4] [Home]

RaptorX-Binding: model-assisted protein binding site prediction for a sequence
[Submit] [Find jobs by: ID or sequence, Email] [Example] [Refs]

RaptorX Contact Prediction: predict inter-residue contacts for a protein sequence
[Submit] [Find jobs by: ID or sequence, Email] [Example] [Refs: 1, 2] [Software Download]

RaptorX Structure Alignment: protein pairwise and/or multiple structure alignment
[Submit] [Retrieve Results by JobID] [Example] [Refs: 1, 2] [Software Download]
Login to your account
Once you submit one job with your email, an account is automatically created for you. Fill in your email address below to see the status of all your jobs.

Server Status
290 jobs pending
188 jobs done in the last 24 hours
4184 jobs done in the last 30 days

#server users: 20105
#processed jobs: 106406

ClustrMaps failed since 09/2015


1. Registration is NOT required, but we STRONGLY RECOMMEND you to provide an email address in submission, which can be used to retrieve the results of all your jobs. Jobs can be submitted using a web browser or a command-line program curl. Each user, identified by an email address, can have up to 500 sequences pending prediction at any time point. Jobs are NOT first come, first-served. Instead, they are scheduled to ensure a fair share of the service. A user has the HIGHEST priority if not being served in the past 2 days. The less served a user is in the past 2 days, the higher priority this user has.

2. A job result can be retrieved by the assigned JobID or its sequence. The result URL will also be sent to the email address if provided in submission, which can also be used to retrieve all your jobs through My Jobs. If you lose the JobID or donot receive the email notification (even in your Spam folder), you may contact the RaptorX team by the form.

3. To see the result web page for interactive visualization, your browser shall support HTML5 and allow JavaScript, which is default on most modern browsers such as FireFox, Chrome, IE, Opera and Safari on several platforms (iPAD3, Windows XP and 7 and Mac). To speed up your web browsers, consult here.

News and updates

System Under Maintenance | 2015-09-30 09:27:50

The computing systems running the servers are under maintenance, so your jobs will be delayed. However, you can still submit new jobs to the servers.

Retrieve Jobs By Their Sequences | 2015-08-12 05:58:42

Sometimes users cannot receive the email notifications upon completion of their jobs due to various reasons. Now users can retrieve jobs by their sequences so that they do not have to keep a record of the job IDs. However, only sequences submitted in the past 60 days can be retrieved since after 60 days it is better to re-predict their sequences, which may result in a better result due to database enlargement.

Contact Server Suspended | 2015-01-31 19:42:06

The server machine running the contact server jobs has a dead hard drive. We will have to install a new hard drive, the operating systems and all the code needed for the contact server, which will take at least one week. However, users can still submit their contact prediction jobs and access the results of previous jobs.