News - Phyre Search - Help - Contact - Disclaimer - Example

DB changes, new hardware, thanks November 19th 2008

I have just moved the entire Phyre system to a much larger computing farm which should speed things up considerably for users. In addition, there have been some fold library changes and sequence database changes. If you have any problems with the new system please let me know as soon as possible. There are bound to be some teething problems at the beginning.

Also I would like to offer my sincere thanks for the large number of glowing letters of support we received for our grant proposal to continue to develop the Phyre server. The grant is now in the hands of the funding body and it will be quite some time (possbly 6 months) before we hear whether we have been successful or not. Once again I would like to express my gratitude to those who took the time to write a letter.

Lawrence Kelley

Delays March 27th 2007

Repeated hardware problems have delayed the intended updates to Phyre.

However, I have decided to list some of the most common requests I have had from users, so at least I can remember what they were.

  • Multiple sequence alignment input
  • Printable version of the results page - I may revert to an 80 column view
  • User added PDB structures and one-to-one threading
  • Library browser to see if a PDB is in the library
  • Reset button on the form
  • Global-local searches in addition to existing local search.

    That's all I can remember right now. If anyone has more things they'd like to see please e-mail me so I can add it to the list

    Lawrence Kelley
  • Database Updates soon - Phyre downtime February 20th 2007

    Some news at last! I finally have the time to do a bit of work on the server. Over the next few days I intend to install a new fold library and get the fully automated fold library updates in place. Because this will involve some code alterations, I will have to take Phyre down for a few days. After that *hopefully* we will have fully automated updates.

    After that, I will hopefully get a few variants of Phyre in an advanced options page, one of them mimicking very closely the behaviour of 3D-PSSM. So those of you who still like the old school methods will be able to use them.

    Finally I hope to get the multiple sequence alignment input in place in the advanced options.

    Thanks for using Phyre!

    Lawrence Kelley
    Database problems - Apologies September 12th 2005
    Apologies for the recent downtime of the server. We experienced some major database crashes. As for good news, the new server has been updated to SCOP 1.69, all PSSMs have been recalculated, and a new supplemental PDB database created.

    There are many improvements on the way, but at the moment I'm in the process of writing a paper on the enhanced version of Phyre.

    Many thanks for your continued patience and support.

    Lawrence Kelley
    User-feedback July 13th 2005
    Many thanks to those of you who have e-mailed me with problems/suggestions/praise. As a result a variety of small changes have taken place, which I think warrants a small version change. Several people have contacted me regarding two issues:

  • Submitting a multiple sequence alignment
  • Aligning a sequence to an uploaded PDB structure.

    I am trying to find the time to implement both of these options. Realistically I give it a few weeks.

    I would also be interested in anyone's experience of the server compared to 3d-pssm. Has Phyre found things 3d-pssm didn't? Or vice versa? Do you find the rather crude functional site prediction useful? Is anyone having problems using Jmol?

    I have also finally fixed the prosite matching. Little blobs appear where prosite hits are found along your sequence and hovering over them will reveal prosite info.

    Once again, many thanks for the feedback and please keep the e-mails coming.

    Lawrence Kelley
  • Beta-testing June 22nd 2005
    The system is now live. A lot of changes are going to happen in the near future. Current improvements over 3d-pssm include:

  • Improved profile-profile recognition algorithm
  • Automated database updates
  • Improved interface
  • Loop modelling
  • First attempts at function prediction

    What to expect in the future:

  • Automated domaining of your sequence
  • Multi-domain assembly
  • Consensus techniques which improve accuracy up to 20%
  • Alignment accuracy improvements
  • Overall and Per-residue predicted model accuracy
  • Ab initio folding options for smaller sequences

    I am currently amusing myself fixing sundry bugs and getting the thing as bomb-proof as possible. Your feedback is particularly useful. If you have any problems/suggestions please let me know.

    Lawrence Kelley
  • Phyre - Completely new protein structure prediction system April 5th 2004
    A brand new fold recognition system is on its way. The new system is nearing completion. Benchmarking suggests it is far superior to 3D-PSSM. However, it requires substantially more CPU power. As a result I will probably have to shut down 3D-PSSM once the new system is up and running. More details to follow.

    Lawrence Kelley

      News - Phyre Search - Help - Contact - Disclaimer - Example

    © Structural Bioinformatics Group, Imperial College, London
    Lawrence Kelley, Riccardo Bennett-Lovsey & Alex Herbert