Introduction

This is a demonstration of a stand-alone replica version of the PDBe Search Database (MSDSD) in mySQL, together with tools and examples about how to use it. All the software together with the demo database is self-contained and wrapped up in a single CR-ROM disc that may be installed and used directly in any Windows or Linux machine. There are no other special pre-requisites apart from 1 GB of free disk space. 

Download MSD-demo CD

This is a cut-down version of the MSDSD in the amount of data that it includes and this was imposed by the need to have a standalone mySQL database of a size that is practical for demonstration purposes (when compressed fits in a CD-ROM). So while the full width of MSDSD is included (100 tables - 1948 columns) instead of the 30695 PDB entries that MSDSD currently includes, this has data for

   - 1254 entries that are included in the SCOP list of representative entries for most tables except
   - 153 entries in the SCOP list with resolution between 1.1 and 1.4 on the residue level (tables RESIDUE, RESIDUE_DATA etc)
   - 49 entries in the SCOP list with resolution between 1.3 and 1.4 on the atom level (tables ATOM, ATOM_DATA etc)

You may get a copy (subject to licensing restrictions) of this MSD-demo CD-ROM disc and follow these steps in order to repeat the PDBe tutorial later. The purpose of this CD-ROM is for demonstration and educational purposes only. Please read carefully the licence.txt file before start using the CD-ROM disc.

This CR-ROM includes
a) Java 1.4 preinstalled for windows and linux
b) mySQL 4.1.12 preinstalled for windows and linux
c) The QueryForm Database Tool
d) Rhino 1.5 from Mozzila
e) mySQL ODBC drivers 02.50.38, setup for windows 
f) Jakarta Tomcat 4.1 servlet container
g) The MSDSD (PDBe search database) a cut-down demo version preloaded in MySQL
h) The MSD-mine servlet web application setup to run locally and access the local mySQL database.
i) A demonstration setup of the MSDSD incremental update mechanism
j) Example files and scripts
k) The standalone static MSDSD data warehouse documentation

 

Ignore steps 1 to 3 if you have already worked out the MSDSD practical session and your mySQL server is already up and running

Contents

  1. Getting started
  2. Configure MSD-demo
  3. Start the MySQL database
  4. Start the MSD-mine web server
  5. Scenario 1.1
  6. Scenario 2.1
  7. Scenario 2.5
  8. Stop the MSD-mine server and the MySQL

Getting started

Install the MSD-demo CD-ROM disc

This may have been done already for you in advance, in order to save time so ask your tutor before you start.
Follow the steps in the install.txt file of the CD-ROM disc and run the install.bat script. This will unzip the msddemo.zip file on drive C:\ and create the C:\msddemo directory.

Test that everything is OK. Open "My Computer" or the Windows "Explorer" and check that C:\msddemo exists and is populated with files and directories

Configure MSD-demo 

Run the script C:\msddemo\bin\config.bat.

This will create a virtual drive O: and will copy the mySQL ini files in the appropriate locations. You will have to rerun this script after each reboot     of the host in order to re-create the virtual drive O:

Test that everything is OK. Open "My Computer" or the Windows "Explorer" and check that the virtual drive O: exists. Also check that the file C:\my.cnf (mySQL configuration file) exists

Start the mySQL database

Run the script O:\bin\startdb.bat. Leave the mySQL server window open 
040114 10:37:02  InnoDB: Started
mysqld: ready for connections

Start the MSD-mine web server

This is the middle-tier of the MSD-mine web application. It operates on your local MSDSD on your local mySQL server and provides a web service that supports the MSD-mine web application. The MSD-mine is a Java servlet that runs inside the Jakarta Tomcat container.

Run O:\bin\startimine.bat. Leave the MSD-mine server window open 

15-Jan-2004 10:40:10 org.apache.commons.modeler.Registry loadRegistry
INFO: Loading registry information
. . .
INFO: Initializing Coyote HTTP/1.1 on port 8080
Starting service Tomcat-Standalone
Apache Tomcat/4.1.24

Now you may access the MSD-mine web-application through a browser like Internet Explorer or Mozzila. Double click on the “O:\bin\MSD-mine.url” URL link file or access the URL http://localhost:8080/imine/servlet/imine.hinterface.FilterBuild from your browser

First MSD-mine example scenario (1.1): Quering

This example demonstrates the MSD-mine querying and filtering and its capabilities of guiding the user in building flexible ad-hoc queries with handy metadata and documentation.

We will get all entries with resolution > 2.0 and having HAEMOGLOBIN or something similar in their title.

We first choose to use the "Structure" mart (section of the database)
And then we choose to use the entity "Entry".













Now we are ready to define our first constrain for "Entry". We want the entries that have a Resolution that is greater than 2.0.
















We also want to put constraints in the Title of the entries, since we want to find heametological interesting proteins. We want as well the title has to contain the work "HAEMO" or "HEMO" or "GLOBIN". Since this is an logical "Or" expression we need to define a sub-expression in order to say that the resolution is > 2.0 and (the title contains the word HAEMO or HEMO or GLOBIN). So we click on the Sub-expression icon

Now we give that we want to logically "Or" the constraints and that Title contains GLOBIN or HEMO or HAEMO.

We then click the execute button

A new window, the "Result window" appears that contains the results of our query. We may click the next page button a couple of times in order to load more records or directly a link to a specific result page. We click "OK" when we our done to stop the query and close the results window .





When done we press the “New Filter” icon on the “Filter build” page to continue with our next example.



















Second MSD-mine example scenario (2.1): Charts – Statistics

This scenario demonstrates the data analysis and charts generation capabilities of MSD-mine. What we will examine is the distribution of assembly types, which means the we will get a chart of the number of monomeric, dimeric, etc assemblies in the PDBe database.

It is important to understand that MSD-mine does is not only accessible from your own workstations but it is in fact a web server that provides its service on the network. Ask the hostname or IP address of the workstation next to you (ask the person next to you to run “hostname” or “ipconfig” on a cmd prompt) and change the “localhost” on your URLs with this. Now you are using his MSD-mine server and MSDSD on mySQL. Now proceed with the example:

Firstly we will choose the "Assembly" entity of the "Structure" mart as the starting entity. Add the constraint that “Assembly Serial != 0” to use real assemblies. We simply then press the "Execute" button to get the list of assemblies in the PDBe database











In the results page we will see the first 40 assemblies and their types.
By clicking on the analyse icon for the "Assembly Type" attribute we will be able to get a chart with its distribution.











We choose to do a complete analysis of the whole PDBe database (not just of the 40 records we have loaded), so will check the "complete analysis" checkbox. We also want to make our chart a bit wider (1000 pixels) so we change the corresponding text box and then we press the analyse button. In the results we see than the analysis was done on the 3698 assemblies that exist on the PDBe database and the chart with the 10 most popular types. By moving the mouse cursor on top of say the dimeric bar we can see that 1248 of these assemblies are dimeric.










Third MSD-mine example scenario (2.5):Combining entities - Advanced analysis

In this example we will demonstrate a more advanced scenario of using MSD-mine, where we will combine (join) entities and perform advanced analysis operations like drill-down and analysing an attribute on the basis of another. We will examine the active site contacts of helices that are part of beta-alpha-beta motifs

We start from the "Motif" entity (of secondary structure mart) and combine it with the "Helix" entity. We further use the "contacts" relationship to find the Residue contacts of the helix. We also choose to filter and retrieve only these contacts where contacts count is greater than three. We then press the "Execute" icon.











After navingating througth the results we choose to analyse the linearity of these helices and we press the analyze icon for the linearity attribute.







We want to do a "complete analysis" of the linearity of the helices of these contacts splitting the whole range of linearity in 20 sub-ranges and get back a line chart.After selecting the appropriate options we press the Analyze icon.

















We decide to focus (drill-down) on an area of interest for the range of linearity between ~ 18-32, for that we click on the corresponding points on the line chart and then we press the analyse button once more.















We want now to examine if there is a relationship between the linearity of the helices of these residue contacts and the bond strength of the contacts, so we decide to examine the average bond strength for different values of linearity. We close the analysis page by clicking the "OK" icon and back on the results page we choose to use linearity as the basis attribute by click the "analysis base" icon for the linearity attribute. We then click the "analyze" icon of the bond strength icon. We choose once more to split the range of linearity in 20 sub-ranges, do a complete analysis and get a line chart. We also choose to use the "average" as the grouping operation. Our chart now has determined the whole range of linearity in our result set, has splitted the range in 20 subranges, has sorted the result records in 20 subsets based on the range that their linearity falls in and has calculated the average bond strength of these 20 subsets.











We see that while the bond strength seems to remain constant and unaffected by the linearity of the helices, there is an area of irreguralitty for linearities between 34-62. So we click on the corresponding icons in the chart and we decide to focus in this area by pressing once more the "Analyse" icon. We then may see that while there were almost 20000 records in the initial result set, there are just 200 in the area or the irregularity. It is easy to understand then that this irregurality is rather a result of statistical error than any indication of a relationship between these 2 factors.











Stop the MSD-mine web server and the mySQL database

Run the script O:\bin\stopimine.bat to stop MSD-mine and O:\bin\stopdb.bat to shutdown mySQL