The batch calculation option enables the analysis of sets of PDB entries.
As a starting point the calculation of mean amino acid distances is currently available.

We plan to add other calculation options, based on user requests.

It is not necessary to keep the Internet browser open until a calculation is finished.

On the batch calculation start page there is an option to view the status of any request.

If permanent cookies are enabled in your browser (at least for the JenaLib server), you can get a history report of all recent batch calculation requests that were started from your browser. Just click on the "go" button without providing a request ID.

If they are not enabled you can still get a status report for a specific request by entering the request ID. To be sure that you can access a request after closing your browser it is best to write down the request ID after starting a request. This will also allow you to view the results in a different browser (on a different computer). The results of a request are usually available for several days after it is finished.

PDB entries are selected for a batch calculation request by providing a list of PDB codes.
One can either write or paste the list into the text input field or select a file containing the list.

If only the PDB code is provided, all available chains are used for the calculation.
If a chain identifier is appended to the PDB code, only this chain is used for the calculation.

A chain identifier can either be appended directly to the PDB code ("1lflA") or by an underscore ("1lfl_A") or by a space character ("1lfl A").

There are 3 different list formats accepted:

• A comma separated list of PDB codes with or without chain identifier.

  example:

  1jnx,1lfl_A,1cft

• A newline separated list of PDB codes with or without chain identifier.

  example:

  1jnx 1lfl_A 1cft

• The list format of the PISCES service (which provides predefined and customized sets of PDB entries).

  example:

  IDs length Exptl. resolution R-factor FreeRvalue 7ODCA 424 XRAY 1.600 0.20 0.23 1WB4A 297 XRAY 1.400 0.11 0.13 2QUDA 125 XRAY 1.600 0.16 0.20

The mean distance between the alpha-carbon atoms of amino acids within a protein chain (not between different chains) is determined for a set of PDB entries.

The distance calculation is performed by using the ncont program of the crystallographic program suite CCP4.

The individual distances provided by ncont are used for averaging over each individual chain and for averaging over all chains of all entries. Averaging (mean distance, median distance) is done for all amino acid pairs or amino acid group pairs that were selected in the submission form.

The amino acid selection can be made in 3 different ways:

1. By an amino acid pair
   One can select either a specific amino acid or "any" amino acid for each half of the pair.
   If "any" is selected, all available standard and non-standard amino acids are included.

   example:

   Asparagine (Asp,D) Glutamine (Gln,Q)

   Averaging is done for each individual amino acid pair.

2. By a predefined amino acid group pair
   One can select predefined sets of amino acids for each half of the pair, according to the amino acid classification by Russell, Betts & Barnes.

   example:

   Aromatic (F,H,W,Y) Charged (D,E,H,K,R)

   Averaging is done for the whole group pair and not for the individual amino acid pairs.

3. By a user-defined amino acid group pair
   One can define a customized group for each half of the pair by supplying a comma-separated list of amino acids. By this way also specific non-standard amino acids (e.g.: MSE) can be included.
   The amino acids can be specified either by three-letter or one-letter codes.

   example:

   Asp,Glu H,L,R

   Averaging is done for the whole group pair and not for the individual amino acid pairs.

The selection of PDB entries for the distance calculation is described in the section PDB Entry Selection above.

Be aware that the mean distance calculation may take several hours or even much longer.

The running time depends on the number of individual distances that must be determined. And this number strongly depends on the following parameters (besides the JenaLib server usage):

• the number of PDB entries
• the number of PDB chains
• the chain length
• the number of amino acid pairs

				Tyr-Leu		Positive-Negative		Any-Any
Code	Entry Count	Chain Count	Residue Count	Time	Distance Count	Time	Distance Count	Time	Distance Count
1bl0	1	1	116	<1 sec	72	<1 sec	330	7 sec	6670
1deh_A	1	1	374	<1 sec	174	<1 sec	1800	18 sec	69751
1deh	1	2	748	<1 sec	348	<1 sec	3600	30 sec	139502
-	100	141	24739	25 sec	13419	38 sec	79503	9 min	2519076
-	2000	3951	837877	11 min	820226	25 min	4455477	7 hours	137350398

The results are available in 2 different formats:

• HTML (HyperText Markup Language)
• CSV (Comma/tab Separated Values).

The HTML result page contains the request status information, a table averaging over all entries and a table averaging chain-specifically.
The individual distances are not included because the page size would become very large, even for a few PDB entries.

In CSV format (tab separated values, suitable for importing the results into spreadsheat programs like Microsoft Excel) there are three independant data files available:

1. averaging over all entries
2. chain-specific averaging
3. individual distances

Because the data files can become very large, we recommend to download them as a compressed ZIP archive. The content of the ZIP archive can be selected freely.