※ Documentation:
Frequently Asked Questions:

1. Q: Since SSP 1.0 is a Delphi Windows program, why not make it available for more than one platform, and preferably as a web server?

     A: What we made SSP 1.0 into an executive Windows program rather than a web server is based on two concerns: 1) Users can download the package and run it in local for convenience; 2) It will promise the security of the sequences data for users.

2. Q: Does this program only predict mouse and human SUMO substrates?

    A: Yes. The identification of the SUMO substrates with their sites in mammlians especially human is more important and urgent recently. And it's hard to gather enought experimental verified data as positive control data-set in other species, such as Fruit Fly or Budding yeast.

3. Q: Is it limited to use of mouse orthologs to support predictions in human proteins? And why was mouse (and mouse only) chosen as the ortholog reference set?

     A: Maybe. But for our work, human-mouse is the right time of separation for this type of analysis. if too close, like human-chimp, many non-functional sequences have no time to diverge. if beyond human-mouse, say human-Zebrafish, it's too diverged to get good ortholog information. And the orthology information between human and mouse is more integrated and abundant than other mammlians.

4. Q: Can this software be used for predicting other than human SUMO substrates?

     A: Sorry, no. SSP 1.0 focus on the substrates and their sites predictions for human and mouse. For other species, users can try other excellent systems/web servers, such as ELM.

5. Q: Is the SUMO pattern in SSP 1.0 too stringent (Ψ-(A, F, I, L, M, V, W)-K-X-E), contrasting the pattern relate to the MOD_SUMO in ELM?

     A: We follow a relatively stringent SUMO pattern to maximum reduce the potential false positives. Surely, There are some reports on the consensus motif/pattern of SUMOs. In Frauke Melchior paper, the consensus motif was proposed as Ψ-K-X-E/D. But rencently in Jacob-S. Seeler and Gill. G paper, it was proposed that the motif of Ψ-K-X-E. Although P is also a hydrophobic amino acid, but in reference such as in Yingming Zhao paper, the P is not included. And in our work, we choose the motif/pattern following the references, without training with existed verified data. And in this paper, our focus is maximum reducing potential false positives with phylogenetic conservation. And the motif/pattern we chose nearly didn't reduce any sensitivity, constrasting to the existed tool SUMOplot.

6. Q: Why does SSP 1.0 use the consensus sequence Ψ-K-X-E, not a longer one?

     A: We used the consensus motif/pattern of SUMO substrates from the recent articles, without training from the existed verified data. Also we collected 67 known SUMO sites in human and mouse with 5 aa before and after the above consensus sequence (14 aa in total), and the logo of them is as the following:

It is obvious that the length of the above consensus sequence is enough.

7. Q: Since you have an extensive dataset, was PSSMs considered as an alternative to patterns?

     A: Yes! PSSMs is a good choice. But in this paper, we focus on reducing potential false positives with phylogenetic conservation, which can be easily combined with the pattern recognition method to improve the efficiency of the prediction. And we chose the motif/pattern from the published references rather than training the data, in oder to avoid the bias for existed data.

8. Q: Why can't we download SSP 1.0 after registration?

     A: The downloading of SSP 1.0 will be permitted after the acceptance of the paper.


The manual of SSP v1.0 can be found in the "Help" window of the software SSP v1.0.

Please register before you can download SSP v1.0, and we will activate your ID
when the periodical updating of this web site...

You can download our supplementary materials of SSP article for your own analysis.

If you use our SSP v1.0 or data for your research, please cite the following article:

Fengfeng Zhou, Yu Xue, Hualei Lu, Guoliang Chen, Xuebiao Yao. A genome-wide analysis of sumoylation-related biological processes and functions in human nucleus. FEBS Lett. 2005 Jun 20;579(16):3369-75. Supplementary materials.

A total dataset of 2,683 potential SUMO substrates in human and mouse identified is freely assessable in txt or pdf format.

Last update: Jun. 19th, 2005
Copyright © 2005-2009 The CUCKOO Workgroup, USTC, All Rights Reserved