Robetta is a protein structure prediction service that is continually evaluated through CAMEO
Features include a relatively fast and accurate deep learning based method, TrRosetta, and an interactive submission interface that allows custom sequence alignments for homology modeling, constraints, local fragments, and more. It can model multi-chain complexes using comparative modeling and provides the option for large scale sampling.
It uses the PDB100 template database, which is updated weekly, a co-evolution based model database (MDB), and also provides the option for custom templates.
Computing resources are provided by the Baker lab and by volunteers from the distributed computing project Rosetta@home. You can help this service by joining Rosetta@home.
Any feedback to help improve this service is greatly appreciated. You can submit feedback and bug reports using the form on our contact page.
Recent alerts and bug fixes:
March 8, 2021 - The option to supply your own multiple sequence alignment for TrRosetta jobs has been added in response to user feedback.
September 15, 2020 - A deep learning based modeling method, TrRosetta, is now available as an option for modeling. For more information on this method, you can refer to the TrRosetta publication.
August 20, 2020 - Last week we suffered a disk failure which affected jobs that were active during the period since the latest data backup. As a result, some jobs may take longer to complete than normal. Sorry for any inconvenience.
July 11, 2020 - We are back online however there is a large queue so please keep in mind that jobs may take some time to complete.
July 10, 2020 - Robetta will be down indefinitely due to technical difficulties with our server. Sorry for any inconvenience.
May 18, 2020 - There was a bug that was recently fixed in the PDB templates detection and alignment method for multi-chain modeling. If you submitted a multi-chain sequence (using a forward slash, '/') without providing your own template, you may want to resubmit your job.
March 18, 2020 - COVID-19 alert. If you plan to submit sequences related to COVID-19 please add "COVID-19" to the "Target Name" of your submission and please keep the job open to the public to aid others who may also be working on the same target(s).
March 2, 2020 - The feature that provides a single 'quick' RosettaCM model from Alignment cluster pages was recently fixed. Please feel free to contact us if you notice any other issues or bugs.
January 31, 2020 - The PDB100 database has not been updated since last December due to an issue. We are working on updating the database and continuing the weekly updates.
January 8, 2020 - Happy New Year! In response to user feedback, we added the option to submit up to 20 jobs. Priority is given to users with no active jobs.
October 22, 2019 - Job run times have been delayed due to filesystem issues. Sorry for any inconvenience.
August 5, 2019 - Job run times have been delayed due to filesystem and database issues. We are working on resolving these issues. Sorry for any inconvenience.
July 2, 2019 - We're experiencing issues with our filesystem and job distribution which have delayed jobs and may have affected single RosettaCM model jobs which were submitted from alignment cluster pages. If you recently submitted such jobs, you may have to resubmit by clicking the "Rosetta Model" link again. Sorry for any inconvenience.
June 5, 2019 - There was an issue with our filesystem which caused a delay in job run times. Sorry for any inconvenience.
May 31, 2019 - Thanks to Luki Goldschmidt for not only building, developing, and maintaining the computing infrastructure behind this service but for also creating the home page graphic. We've added the total number of models sampled per week along with the previously existing usage statistics displayed below.
May 1, 2019 - There was a bug in our contact page. Feel free to send us feedback.
April 11, 2019 - You can now view non-private jobs without having to log in. Fixed a bug on the submit page when trying to add constraints without a loaded template.
March 27, 2019 - Updated the PDB template upload reader to skip residues missing backbone atoms which cause Rosetta jobs to fail. When the input sequence is modified and in the rare case when the template(s) realignment fails, the templates will be removed.
March 22, 2019 - Fixed user provided partial thread input error where coordinates with no chain identifier would get skipped. Changed how multiple user provided partial threads were setup for RosettaCM. Previously they were structurally clustered and each cluster would be modeled separately. This was changed so multiple user provided partial threads are no longer clustered. All custom partial threads are now used in a single RosettaCM hybridize run.
PDB templates last updated April 01 2021
Jobs queued: 0
New users last week: 188
New jobs last week: 2280
New final models last week: 10044
Total models sampled last week: 9391195