Why is CHRAB needed? - you might ask. "Doesn't existing tools like Blast, SMART, Pfam and InterPro give such information already?" Sure, these are indispensable tools for finding domains and similarities among proteins. It is our experience, however, that currently available tools do not give sufficient detail in the prediction and description of protein sequences. For instance, you might find that your protein contains a chromo domain. Current literature will then tell you that some chromo domains bind methylated lysines on histone tails while others bind RNA. How do you determine which function is most likely for your protein? Another example is the SANT domain. Some tools will tell you that your protein has a SANT domain, while another tool reports a myb domain, known to bind DNA. What do you do then? CHRAB is intended to help resolve problems by bringing more detail to protein domain prediction.

"How can CHRAB do that?" - CHRAB will be based on an exhaustive re-evaluation and re-classification of proteins and protein domains involved in chromatin-based gene regulation. CHRAB will employ several different prediction methods and combine this information knowledge from experimental data on individual proteins and protein domains in order to give more detail to the predictions. Our approach will be, in part, similar to that used for re-evaluating the RING fingers of Arabidopsis (Kosarev et al. 2002 [GB:online]).

"When will CHRAB be available?" - A protype CHRAB server including an HMMer server for detection of chromo and SET domain is now in operation. Diagrams which illustrate the CHRAB data model are available here [pdf].

If you have questions or ideas about how CHRAB could work, don't hesitate to contact us at chrab@uib.no. In particular, if you would consider helping us recording information and knowledge about chromatin-associated proteins or protein domains, please contact us. We will soon operate a web-based interface for recording experimental data.

You can also visit our .... somewhat outdated ... webpages for the PHD finger, Chromo shadow domain and the SANT domain. Check out our new data on the nucleosome-binding properties of the PHD finger now in Ragvin et al., J Mol Biol. 2004 337(4):773-788.

Project partners and collaborators:
Rein Aasland, Department of Molecular Biology, University of Bergen
Katharina Reksten Tufteland, PhD-student (NFR; 6/02)
Jan Christian Bryne , MS-student (II; 1/05)
Grigori Merkine, Exchange-student (NFR; 1/04)
Pål Puntervoll, post-doc (CBU)
Karin Wibrand, Scientist (on CHRAB til 2003)
Reidunn Aalen, Department of Biology, University of Oslo.
Andrew Lambertsson, Department of Biology, University of Oslo.
Inge Jonassen, Department of Informatics & CBU, University of Bergen.
Gisle Sælensminde, CBU, University of Bergen.
Carsten Helgesen, Bergen University College / (Meltzer; 8/02 25%)
Francis Stewart, Technical University of Dresden.