1. What is a "Template Search"?
2. What is the "Search via Known Compounds"?
3. What is the "Search via Supplier/ID"?
4. What is the "Search via your own Substructure"?
5. What is the "Search for Similar Drugs"?
6. What is the "Search for Cellular Effects"?
7. How can I perfom a 2D Similarity Search against the database?
8. How can I perform a 3D Superposition?
9. How does a 3D superposition work?
10. How long takes a 3D superposition?
11. How can I save two superimposed structures?
12. What are the System Requirements for MDLŽ Chime for MDL® Chime?
13. How are the conformers generated?
14. How is the database generated?
The Template Search allows you to search the database with a small assortment of substructures. Only such substructures are available, which can be found in the database and where exist no more than 450 hits per substructure.
The search via known compounds allows you to
search in a subset about 2500 common natural compounds. It is also possible to
search for a compound of interest by a variety of methods. Ranging from simple
search methods like the name or the molecular weight up to sophisticated search
criteria like the number of chiral centers. Most of the substances are also searchable by their classification which
can be selected from a pulldown menu.
With this type of search you can find similar structures of known compounds,
which are available by 8 different suppliers.
The search via Supplier/ID allows you to
search in all natural compounds available in the Super Natural Database. To specify your search simple
search criteria like the ID or the name of the supplier can be selected as well as advancerd search
criteria like the number of chiral centers or logp.
With this type of search you can find similar structures,
which are available by 8 different suppliers.
The search via marvin applet allows you to
build your own substructure and perform a similarity search in all purchasable natural
compounds.It is also possible to load different molecular structure files into the applet
and start a similarity search.
With this type of search you can find similar structures,
which are available by 8 different suppliers.
The search for similar drugs allows you to search in the free drug database ( SuperDrug Database) for similar compounds.
The search for cellular effects allows you to search a compendium of the SuperNatural Database for their activity in various tumor cell lines. The National Cancer Institute (NCI) has screened about 60,000 compounds in a cell panel of about 60 tumor cell lines since the 90s. For all compounds the 50-percent growth-inhibitory concentration (GI50) is given as an index for cytotoxicity and cytostasis. We added the GI50 profiles of more than 400 natural compounds from this screening program to the SuperNatural Database. The search for cellular effects enables the screening via NSC number (NCI ID), a variety of tumor types (breast, CNS, colon), the cell line panel (DU145, HCT-116, MCF7) and via a specific GI50 range. The results display the GI50 values for each cell line, which was treated with this compound. Furthermore, the supplier for this compound is indicated and the user can perform similarity searches to reach structural similar and most likely similar active compounds within the SuperNatural Database. The total growth inhibition value (TGI) and the lethal concentration value (LC50) for each compound will be added soon.
Starting from a certain compound, the Super Natural Database can be searched for similar molecules. A 2D similarity search is performed by calculating Tanimoto coefficients for the given molecule and the molecules in the database.
After you have performed a compound- or template - search, simply click on the Search-Button in the "Similarity-Search" column of the search results table. You can also select how many hits you want to see.
Work of 3D Superposition
Three-dimensional superposition of two different molecules is performed in the following way: each conformation of one molecule is superposed with each conformeration of the second molecule. Those two instances matching best are displayed. The best match is defined by maximizing the following score:
score = [(Number of superposed atoms) / (Number of atoms in the smaller molecule * e-RMSD) ]
where RMSD is the Root Mean Square Deviation of the superposed atoms.
It depends on the server loading time and the number of conformers for both compounds (max.20-30 sec). If a superposition does not succeed, it may be due to server overload. Please try it again a few minutes later
Save two superimposed structures
Learn more about how to use Chime®: Video 1 (1.4 MB), Video 2 (3.9 MB)
System Requirements for MDL® Chime
PC
Software: for Chime 2.6 SP5
Hardware:
Macintosh
Software:
Hardware:
For more detailed information please visit MDLŽ Chime.
We generated the conformers with "Catalyst 4.7", a software from Accelrys. This program uses a poling algorithm to promote conformational variation. The poling algorithm helps to eliminate much of the redundancy in conformer generation and improves the coverage of conformational space. To remove similar conformers Catalyst screens the generated conformers to see if any two conformers are within the RMS tolerance. The RMS tolerance is computed as:
RMS_TOL = 0.1 +C * sqrt ( 1 + Nrot )
where C is a scaling constant and Nrot is the number of rotatable bonds (allowing the tolerance to scale with molecular flexibility). All conformers in the Super Natural Database are generated with the default value of parameters.
Catalyst can use two different methods of conformer generation. FAST and BEST. For the Super Natural Database we used the FAST-method, which is proposed from Accelrys for building databases.
Another parameter of Catalyst is the maximum energy for each conformer. We take a threshold of 20 kcal/mol as a relative maximum energy.
Catalyst allows you to define the maximum number of conformers for each compound. Values from zero until open end are possible. We generated all conformers with a maximal number of 100 conformers for each substance.
In Catalyst the chiral centers can be set as absolute (the stereocenter is known) or unknown. This has a direct impact on the conformer-generation routines, because if a stereocenter is marked as unknown it can be either R or S. Catalyst is able to read the chirality flag in the MDL MOL-Files. For our conformers we set the stereocenters absolute, provided that the stereochemistry is known.
If you want to get more information about the conformer generation you can have a look at this description from Accelrys.
If you want to get more information about the database architecture you can have a look at this Schema.