The Database of Quantitative Cellular Signaling is meant as a community resource for modeling signaling pathways at the level of chemical reactions. It is intended to make it easy to extract reaction schemes and parameters for published models and to be an archive of signaling models at various levels of detail and various stages of model development. It is also intended to provide a forum where experimentalists can comment on model properties, and modelers can appeal to experimentalists for specific data.

Most model parameters in the database are from public sources such as published papers and freely available models. All parameters are made freely available with the request that users acknowledge the database and where appropriate, the specific data sources cited in the database. Users of the database are permitted to copy the parameters provided they remain freely available. Parameters are made available without any form of express or implied warranty about their utility. Licensing is under the terms of GNU Public Licence (GPL).

The database is copyrighted (C) 2001 Upinder S. Bhalla and the National Centre for Biological Sciences (NCBS), Bangalore, India.
NCBS is a centre of the Tata Institute of Fundamental Research, Mumbai.
The copyright on the database does not imply copyright of the original data sources or models, which remain the copyright of their respective owners or in the public domain, as applicable. The source code for the database is also
copyrighted (C) 2001 Upinder S. Bhalla and the National Centre for Biological Sciences.
The source code may be requested from the author and is licensed under the terms of the GNU Public License Version 2. As mentioned above, the copyright terms are intended to ensure continued free availability of the data in this database.

The parameters are obtained from experimental sources such as binding experiments, time-courses of reactions, enzyme assays, immunoblots and just about any other experimental technique that sheds light on the pathway chemistry. Modelers first try to work out a simple chemical reaction scheme that embodies the interactions which they wish to examine. Then each of the parameters in this reaction scheme must be filled in from one experimental source or another. Often the reaction scheme itself must be re-examined on the basis of the available data. A more detailed description of this process is presented in a couple of book chapters:

Bhalla, U.S.
Simulations of biochemical signaling
Computational Neuroscience: Realistic Modeling for Experimentalists. Ed. E. De Schutter. CRC press, 2000

Bhalla, U.S.
Use of Kinetikit and GENESIS for modeling
Methods in Enzymology. Volume 345 Ed. R. Iyengar and J.D. Hildebrandt. Academic Press. 2002

Each model (and often each parameter) has its author's assessment of its accuracy, as well as comments from outside experts. The short answer is that most models are OK to within about a factor of 2 in their domain. The long answer is to test out the model and see what it incorporates and what it leaves out. Many models include comments about sensitivity analysis.

As models are always an approximation to the biological system, it is particularly important to have an evaluation of model applicability. This is done through a curation system, whereby modelers and experimentalists are invited to comment on specific pathway models. Curation entries receive an accession number.

As with any web project, this is work in progress. At the current time your best bet is to email the database maintainer, Upinder S .Bhalla bhalla@ncbs.res.in, who will figure out how to format and enter your pathway. Very soon we hope to have on-line data entry facilities.

Pathway entry will remain moderated for the foreseeable future. On-line entries will automatically receive a tentative accession number, but the entry will be examined to check that it is not frivolous.

Initially, all contributions are identified simply by an accession number. The longer-term plan is to identify all contributions to this database by accession identifiers which convey a little more information, as follows:
The base accession number being the name of the signaling pathway family.
The second field is the accession type, currently one of

• Original
• Network
• Curation
• Parameter update
• Interaction
• Subtype
• Derived model
• Spatial/distribution information

For example, an original model of Gq might have the number

Gq_Orig_1234

PLCbeta_Deriv_4567

A Model is description of a pathway or a network of pathways described in GENESIS scripting language.

An Accession list is just what the name says. It is the listing of models according to their unique Accession number.

An Accession type defines whether the accession number is for a pathway model or a network model.

A pathway is a sequence of reactions which are generally considered to be a complete unit e.g. the MAPK cascade

A network is a group of interconnected pathways that are a functional unit. E.g. A synaptic signaling model described on DOQCS has PKC, MAPK, PKA and CaMKII pathways.

A Pathway list is just what the name says. It is the listing of individual pathways that are participants of network models listed according to their unique Pathway number.

Accession Details state the name and model type of a particular accession.
Details about the model annotation are available in accession details. Also the available model formats for the particular model are listed here.

Pathway Details state the pathway identifiers like the name and number of the pathway; name and number of the accession of which the pathway is part of.
Details about the pathway annotation are available here. Also the available model formats for the particular pathway are listed here.

Related Pathways are a listing of pathways similar to each other. The similarity is based purely on the names of the pathways. This listing is available in the Pathway Details.

The Shared_Object_pathway is the collection of molecules, enzymes and reactions that occur in more than one pathway within a network model. These are responsible for communication between the pathways in the network.

Some molecules, enzymes and reactions shown in the reaction scheme diagram for a pathway in older network models, might not be listed in the respective lists for the pathway. These are probably a part of the Shared Object and would be listed under Shared_Object_pathway

The majority of biological pathways represented on DOQCS are neuronal signaling pathways. There are signaling pathways from other systems like MAPK pathway in Xenopus oocytes, E.coli chemotaxis pathway, Firefly luciferase pathway, cardiac pathways among others.

An entry date is the date on which the model was deposited on DOQCS.

A Developer is the original creator of the model in its original format cited in a publication.
A Transcriber transliterates whenever possible the original model format into the GENESIS format for Kinetikit which is available on DOQCS. Transliterated models match the original published model in all parameters. Sometimes when converting original model formats to GENESIS better matches to published literature would be possible by using different parameters from original model publication. Modified parameters are obtained by curve fitting or by personal correspondence with the developer of the model. Such translated models always contain a note specifying the changes made from the original published models.

A Kinetikit file or a .g file is a text file which contains the model description in the GENESIS scripting language that can be "read" by Kinetikit to simulate the model. These files end with the .g extension.

The .g file has the information required by Kinetikit to create the model for simulation.

The .g file can be downloaded from either the accession entry or pathway entry for a model.

The acc.g file is the .g file for the composite network model with the parameters set for the specific network behavior. The path.g file has parameters set for specific participating pathways within the composite network.

Sometimes there are no separate files for the accession and the pathways therein. This is because both system behavior of the network and the behavior of the participating pathways can be seen through the same .g file.

The user can run a .g model file by downloading the .g file, Kinetikit interface and GENESIS. Detailed instructions for running the model are available here.

Kinetikit and GENESIS can be downloaded from here.
Size: ~1.2 MB download and 4 MB uncompressed
Requirements: The software requires a PC with Linux operating system (Kernel 2.4 or later, for example, Red Hat Linux 7.2 and higher).

The connected to table lists how the query term is related to other entries in the database.

The Enzyme Type refers to the mathematical formulation used for the enzyme.
The Michaelis-Menten reaction scheme is E + S <=====> E.S -----> E + P
In the explicit representation, each of these reactions is computed and there is an intermediate enzyme-substrate complex E.S which is formed. This formulation is somewhat more chemically complete. Note that it is still an approximation as no backward reaction is permitted for product formation.
In the classical representation, the computation of rate (i.e, rate of formation of P) is done using the equation rate = [E][S].kcat/(Km + [S])
where [E] is the total amount of free enzyme. These two forms usually behave similarly, but give different effects when there is enzyme saturation and more than one substrate is present. Other, more complex enzymatic mechanisms are well known, but can be represented as assemblies of simple reactions.

An accession name is essentially the name of a model on DOQCS. Default accession searches on DOQCS are done in the name field and the results are not exact word matches; e.g querying for CaMKII in accession list will yield fig3_CaMKII, CaMKII and CaMKII_2003 among the search results.
Osc_Ca_IP3metabolism is a model on DOQCS that has CaMKII as one of its participating molecules. But this is not returned as a result for CaMKII accession search because the model name is "Osc_Ca_IP3metabolism" and does not have CaMKII in its name. The Model serach in Connected to listing yields all accessions that have CaMKII as a participating molecule.
In short
Accession name results: models that have the queried term in the accession name.
Connected to model search results: models that have the queried term anywhere in the model, not necessarily in the accession name.

You may contact Dr.Upinder S. Bhalla at this email ID bhalla@ncbs.res.in or
HarshaRani at this email ID hrani@ncbs.res.in