2013-02-01
* Second major upgrade of the server.
* One more nucleus added: CheShift-2 computes also 13Cβ chemical shifts values.
* Report of sidechains torsional values: For residues with large differences between computed and observed chemical shifts, the server tries to provided alternative rotamers to decrease that differences. This feature is only available if both 13Cα and 13Cβ chemical shifts values are provided.
* Outliers: Observed chemical shifts with unusual values (according to the BMRB database) are reported, also residues located in low probabilities zones in the ramachandran map.
* Chemical Shifts for cysteine residues are no longer computed: We are working on increasing the accuracy of the computations for this type of residue.
* Differences between observed and computed chemical shifts values are no longer smoothed.
* New color code: Now green, yellow and red colors represent small, medium and large differences, respectively. Blue is used to mark residues that could improve the agreement between observed and predicted chemical shifts for both the 13Cα and 13Cβ nuclei, if the sidechain torsional angles are changed. White is used if either the prediction failed or the observed value is missing.
* CheShift-2 back-end rewrited: The code has been ported to Python. Numpy and Scipy are used for the cpu-demanding tasks. The code is now ~10 times faster.
* No more Java-plugin needed: JSmol is used now. JSmol is a Java-free version on the Jmol applet using only HTML5. If for some reason the browser does not support HTML5 technology JSmol will try to automatically use a Java to display the protein.
* Minor corrections to the computations of all residues: statistical corrections were made to increase the accuracy of the chemical shifts for all residues.
2012-03-31
* First major upgrade of CheShift: Now the server is named CheShift-2.
* CheShift-2: incorporation of a graphic validation option on a per residue base.(if the user provided the observed 13Cα chemical shifts).
* Corrections to the computation of some amino acids: Increased accuracy to the computation of residues preceding a proline residue and increased accuracy in the computation of histidine residues.
* Change in the way the results are retrieved: The results are not longer e-mailed to the user, instead results are displayed on the screen.
* CheShift PyMOL plug-in: CheShift-2 can be also accessed from PyMOL using a plug-in available here.
* Increased number of conformations analyzed: The server handle up to 20 conformations.
2009-09-03
* First available version of CheShift: The server predicts 13Cα chemical shifts values.