RCSB PDB adds customized widgets to their homepage.
(if you are having trouble seeing this video try the full screen option)
BRENDA is a gold mine for those studying enzymes! The database proclaims to be the comprehensive enzyme information system and with 5010 enzymes it looks to be the case. Here is a screenshot of the navigation bar. As you can see BRENDA brings together many different categories such as IC50 values, pH stability range and crystallization.
My only suggestion so far is to change ‘Recommended Name’ to ‘Enzyme name’. I think it would save some confusion in the search entry.
I have never seen another database bring together this much information about a class of proteins. If you have a colleague working in enzymology this is site is definitely worth passing along.
Carbohydrates (glycans) are a major class of biological macromolecules along with proteins and DNA molecules. The array of possible combination of carbohydrates is astounding. Here is an extensive list of carbohydrate databases that offer a number of different searching methods as well as entries, enjoy.
1) GlycomeDB is a carbohydrate structure metadatabase that has combined all free databases (CFG, KEGG (right panel), GLYCOSCIENCES.de (no $), BCSDB, GlycoBase (Dublin, Lille) and Carbbank) including both their structures and annotations.
2) Glycoconjugate Data Bank has a really cool search GUI (see ‘Search glycan structures’)
3) O-GlycBase contains 242 glycoprotein entries of both O- and C-glycosylated proteins.
4) Glycan Binding Proteins seems heavily dependent on knowing the CFG ID of your glycan of interest.
5) GLYCO3D breaks down glycans into seperate databases such as monosaccharides, di, oligo, poly, lectins, glycosyltranserases and Gag binding proteins. The search is organized by a series of drop down menus, which is really helpful if you are taking a top down approach (animal to glycan). On the downside, the website has not been updated since May of 2007.
6) Lectins contains lectins from plant, algae, virus, animal, bacteria, yeast and fungus.
7) Carbohydrate-Active enZYmes Database describes the families of structurally-related catalytic and carbohydrate-binding modules (or functional domains) of enzymes that degrade, modify, or create glycosidic bonds.
Glyco Enzymes is good if you have a monosaccharide and want to find out the glyco-enzymes involved. This database is not for finding structural information such as what is found in the PDB.
9) Bacterial Polysaccharide Gene Database doesn’t seem to be working anymore
For structural information concerning metals and RNA then you may find MeRNA quite helpful (pdf ref). Metals have been shown to play an important role in RNA folding. The MeRNA database currently contains 398 PDB entries which include 22 different metals.
Note: The MESPEUS database is excellent resource for metals and proteins.
The simple periodic table display works well. The results page would be easier to read if the titles were not all in italics.
The advance search offers a number of function with my favorite being the ability to search through 8 different binding motifs. The advanced search also allows for search by reference and/or author.
Tableau allows for the searching of protein folding patterns of substructures in the PDB structural database. This type of searching can be helpful in understanding protein structure, function and evolution.
The server searches using secondary structure elements and is capable of finding either an entire structure or a substructure of a larger structure (ref).
Tableau has a number of simple inputs (title, email, desired structure, output), however, you want to read the suggested tips. You can submit structures that contain multiple chains or domains, but it is not recommended. I submitted a number of structures and had a response time of about 3 minutes.
The server is also handy for tracking down fake structures.
The inability of a protein to be crystallized may be due to disorder regions. A work around to this problem is to truncate the protein. The question then becomes where should these truncations be made? A number of prediction servers have been created to address this problem (Nir put together a nice list here).
The metaPrDOS (meta Protein DisOrder prediction System) is convenient in that it predicts natively disordered regions of a protein chain from its amino acid sequence by seven utilizing
independent predictors (PrDOS, DISOPRED2, DisEMBL, DISPROT (VSL2P), DISpro, IUpred and POODLE-S) (pdf: ref).
The ability of this system led to a win at CASP7. (Does anyone ever lose at CASP?)
The inputs are straight forward:
A sample output can be found here (scroll down a little). I have been waiting almost 5 hours for an output, but still no luck.
Have you used any of these prediction servers? Did you like the results? What else are you using to decide where to truncate a protein?
If you are struggling with refolding your protein then you may want to take a look at REFOLD. The Refold database currently contains 759 protein entries.
The help pages are really nice in that they explain how to use the database and contain background information. For example, here is a pdf of ‘A practical guide to protein expression and refolding from inclusion bodies.’ Based on the current entries the most common method of refolding is by dilution.
Here is a shot of a couple of search options (note: all the panels are not open)
Graphs are updated nightly to reflect the current contents of the database:
A neat feature of this database is that it allows users to create their own free account which will track their past searches.
The process in macromolecular crystallography for generating heavy atom derivatives can be tedious. Problems may arise from heavy atoms not being incorporated into your protein to difficulty in producing crystals for derivatization trials therefore making each attempt critical.
The Heavy-Atom Database System: HATODAS II has been created to address these problems. The database uses 93 known heavy atom binding motifs (derived from 3103 heavy atom binding sites) and can take into account the amino acid sequence as well as the crystallization condition (ref).
Here is an example of a prediction that HASTODAS generates for potential heavy-atom reagents:
The following is a list of the suggested motifs that are present in the submitted sequence:
If your protein does not contain a His, Cys or Met then you maybe forced to mutate a residue for derivatization, but which one do you choose? HASTODAS addresses this question by suggesting a point mutation(s) based on multiple sequence alignments of homologous proteins.
Points for creating a database with guts.
Dear Protein Data Bank,
It’s not you, it’s me.
We’ve been inseparable for what seems like forever, we have been through a lot. Unfortunately, I don’t think that our relationship is going to work out.
I’ve done my best to be patient and even offered suggestions on how we could make things better. I know that you have been improving and even updated your site. I just feel that I need to be better connected to other resources.
Maybe I’m giving up on you too soon.
I’ll miss you,
Sean
P.S. I thought you should know that I’ve been seeing PDBsum lately.
I thought the enzyme catalytic mechanisms (ECM) database would be a nice follow up from yesterday.
The amazing part about this database is not the number of entries (720), but in the details. The ECM has devised a classification of enzymatic reactions which are as follows:
R: Basic Reaction
L: Ligand group involved in catalysis
C: Catalysis type
P: Residues/cofactors located on Proteins.
This classification system creates a hierarchy that is then search able by the user. The hydrolysis classification even has pictures of the general mechanisms. Here is an example of a Pepsin-like mechanism and Trypsin-like mechanism. If you or your students are learning general mechanisms a number of these illustrations could serve as excellent real world examples.
The search page contains a number of unique inputs.
For example, ECM utilizes the KEGG pathway database to generate a reference pathway.
Note: use the bottom search button not the top, which is (above the fold) if you do not have a DB code input.