POP-OUT | CLOSE
 

October 05, 2010 Release

 
Table of Contents:
 
  Molecule of the Month Improvements
  PDBMobile for the iPhone
  Query Result Browser Improvements
  Chemical Components
  Tabular Report Improvements
  Comparison Tool Improvements
  General Site Improvements
 
 
Molecule of the Month Improvements
 
With 130 articles as of October 2010, the Molecule of the Month series begins to provide a structural view of biology. To improve the ability to access all of these articles, the navigation of the Molecule of the Month archive has been organized in two new ways:
  • the first way is new category view
  • the second way is the improved table view

New Category View

The category view groups the columns into six major categories: Protein Synthesis, Enzymes, Health and Disease, Biological Energy, Infrastructure and Communication, and Biotechnology and Nanotechnology. Users then drill down into various subcategories to find all relevant Molecule of the Month features.

For example, Protein Synthesis has subcategories for Structure of Nucleic Acids, DNA Maintenance, Replication, Transcription, Translation, and Protein Folding, Modification, and Degradation. The Structure of Nucleic Acids subcategory links to the Molecule of the Month features on Designed DNA Crystal, DNA, and Transfer RNA. Quick previews of the MotM features are available as a mouse over for each of the drill down links. Several articles cover multiple categories/subcategories and are listed under each drill-down.

To access it, click on the 'Molecule of the Month' link under Education in the left hand menu, or the 'MotM Category View' link on the home page.

Show Molecule of the Month Category View

 
 
 

Molecule of the Month Archive

Molecule of the Month articles have also been organized into a table sortable by Category, Date and Title. Each of the listed features also includes a quick preview box that displays the related image for the MotM feature as well as a synopsis. The archive view can be accessed from the 'Previous Features: MotM' link on the home page, or from the category view of the Molecule of the Month.

Show Molecule of the Month Archive View

 
 
 
 
PDBMobile for the iPhone
 

Download Application

NOTE: This is a new software release; we welcome your feedback regarding the application.

PDBMobile is an HTML5-based application for mobile devices. It initially targets iPhone and iPod devices running the iOS4 operating system version 4.1. Future releases will target other platforms, including Google Android devices, Windows Phone 7 and Meego platforms. PDBMobile will also support both 9.7 inch and smaller iPad form factors, and likewise for other small form factor tablet devices. PDBMobile will take advantage of the larger display area to enable concurrent display of search results and PDB entry information screen.

PDBMobile should be installed when on a WiFi network since the initial index download will take a few minutes. After that, the application may be used via WiFi or 3G data connectivity; it has been optimized to provide fast data access on-the-go.

To install the software, first make sure that your iPhone or iPod is updated with the latest operating system (at least iOS 4.1). Then point the Safari web browser in your iPhone or iPod to the RCSB PDB website (http://www.pdb.org), and select the "yes" option to begin installation.

You will be prompted by the web browser to allow the creation of a client-side database; this database is used by the PDBMobile application for index storage and other data persistence.

Detailed installation instructions can be found in the installation screencast and further information about PDBMobile can be found in the PDBMobile FAQ page.

 
 
 

Tapping the "Search" icon in the toolbar (at the bottom of the screen) opens the PDBMobile Search page. Here you can make keyword, PDB ID, author and annotation searches, in addition to viewing lists of the most recent structures, annotated and flagged entries.

Once a search term is entered and the button pressed, the results browser panel opens automatically to display the matching structures. Scroll down to view results; tap the "Next 50" button at the bottom of the screen to display the next 50 entries returned from the query.

Tapping on any entry opens the abstract display panel, which contains the primary citation abstract (where available) and a larger image.

Flipping over the abstract pane by clicking on the note icon displays the annotation text area.

More detailed usage instructions can be found in the PDBMobile screencast.

 
 
 

View the Latest News & Molecule of the Month

From the home page of the application (you can always reach the home page by tapping the home icon on the toolbar) you can access the latest RCSB PDB news items and also the latest molecule of the month.

The four most recent news items are displayed on the info pane when the news ticker is cliked on. Clicking on links in news items will open the Safari browser outside of the PDBMobile application.

The latest Molecule of the Month is also accessible through the home page of PDBMobile. This is a specially formatted version of the iPhone which substitutes the interactive JMol for an interactive image that the user can click through. Structure ID links and the "Related Structures" links displays further information.

We are in the process of working to make all of the Molecule of the Month archive accessible by mobile devices such as the iPhone.

 
 
 

MyPDB

Save Queries

Click on the MyPDB logo to log into your MyPDB account. A listing of any saved queries you have stored in your MyPDB account will be displayed. You can save queries run in your PDBMobile application by pressing the "Save" button on the top right of the query results browser screen. You'll be prompted for a descriptive query name under which to store the query in your MyPDB account.

 
 
 

Personal Structure Annotations

A new feature of MyPDB that is being tested with PDBMobile and will be rolled out to the main website MyPDB interface in the next release is the ability to store personal annotations for any entry in the Protein Data Bank. Annotations are made by flipping over the entry abstract pane and typing into the text area. If the "Save" button is pressed, the annotation will be saved both locally and into your MyPDB account (if logged in). Similarly, flagged entries (by clicking on the small translucent star next to each entry in the query results browser) will also appear in your MyPDB account.

 
 
 
 
 
Query Result Browser Improvements
 

Data Distribution by Category for Query Results

This "Drill-down" feature, which provides data distribution summaries by category for query results, highlights the main characteristics of the PDB entries that match a query to help focus on subsets of interest (more details ...).

For example using the top bar text box and the associated drop-down, we can search for all the PDB structures of author "Ramakrishnan, V."

The Query Results page displays charts summarizing characteristics of PDB entries like resolution, release date and organism.

Clicking on these icons will reveal more detailed summary information and links that can be used to refine results with a single mouse click.

Moving the mouse cursor over the drill-down links and the small chart icon will pop-up additional more detailed charts. Clicking again on the chart icons will display a larger image of the data distribution for this category.

The "Drill-down" feature (also called faceted search) is designed to provide a small number of the major drill-down options. Groups of small size are not displayed to make them more readable.

Drilling down by clicking on a link will either present more granular drill-down options, or a "grayed out" label that indicates that further drill down is not possible since all results belong in a single group.

Note: For categories of numbers or dates like resolution or release date, the drill-down links include the lower bound value but do not include the upper bound value. So for example the drill-down of 1.5-2.0 Å, will include the entries with exactly 1.5 resolution but not the ones with exactly 2.0 resolution (these will be included in the next 2.0-2.5 group)

 

Examples:

Searching for all the structures of author "Ramakrishnan, V." returns more than 100 entries: 77 or so (exact numbers may be different in future updates) are non pure protein (Mixed) and ~76 of them are actually protein/RNA complexes. Clicking on the relevant drill-down link shows that 2 entries are from Electron Microscopy; clicking on the X-RAY drill-down link will refine the query to the 74 X-RAY entries. Finally, we can see that 51 of these were released after 2005, which we can get by clicking again on the relevant drill-down link.

If the available drill-down options are not what we exactly need, we can click on the alternative "more choices ..." link and refine our query more precisely (for example 69 entries that were done after 2000).

It has to be noted that the "Drill-down" feature is available for all types of queries, for any size of result set, and it is particularly convenient for large results sets where navigating over results is not practical.

For example searching for "All alpha proteins" based on the SCOP classification, will reveal more than 7600 results but we can immediately see that just 238 or so are virus structures (by looking at the "Taxonomy" drill-down).

Finally a sequence query for the sequence of protein chain 1FCC:A will reveal that there are about 15 other entries with similar sequence and a resolution that is a lot better (less than 1.5 Å) which we can retrieve by using the drill-down link.

 
 
 

Data Distribution by Category for Entire PDB

The drill-down feature is an easy way to explore the whole PDB archive and the most recently released entries without any explicit searches.

By just clicking on the links on the top of the home page that show the total number of PDB entries and the date of the last update, one may retrieve query results for the whole set of the PDB or the latest entries from the weekly update.

As an example, click on the total number of entries and drill down on the human ones (Organism is "Homo sapiens"), and then drill down further on the ones that also contain a viral part (Taxonomy is Virus), to get all PDB structures with a human and a viral component.

Show Data Distribution for Entire PDB
Show Data Distribution for Latest PDB Entries

 
 
 

Drill-Down by Categories

  • Resolution (in Ångstroms)
    Initial groups are: less than 1.5, 1.5-2.0, 2.0-2.5, 2.5-3.0 and more than 3 Å. If the user drills down inside a group (all the results belong to a single group), then there will be at most 6 groups of almost equal size (1 decimal digit).
  • Release Date
    Groups are: before 2000, 2000-2005, 2005-2010, 2010 and later. Drilling down on one of them presents yearly groups (decade groups for before 2000) and drilling down further presents monthly groups.
  • Experimental Method
    Groups are the types of experiments that determined the structures (X-RAY, NMR, etc.)
  • Polymer type
    Complexes with polymers of different types are presented separately in a "Mixed" subcategory. The "drill-down" link "Protein" will return pure protein entries.
  • Organism
    The exact organisms/strains assigned to a PDB entry are presented. Drilling down on the "Escherichia coli" link for example will not result in entries of E. coli strains (otherwise these entries would be counted twice in the charts). The "Other" drill-down link can be used to exclude the presented groups from the results, and allows navigation over organisms.
    On the root level, organism groups include cases of hybrid entries (cases where an entry has a polymer chain of the selected organism and a polymer chain from another organism). But when the user drills down on a single organism (all results include that organism) the hybrid entries are listed explicitly.
    For instance the initial grouping may display "Homo sapiens (10)" but when the user drills down, he may get "Homo sapiens (8)", "Homo sapiens / E. coli (2)". In the second drill-down, the "Homo sapiens" link will include the entries that are pure human
  • Taxonomy
    Gives the taxonomic group of an entry: Eukaryota, Bacteria, Viruses, Archaea, Other. As for hybrid entries, the root drill-down will not display hybrid cases explicitly.
  • Sequence length
    This is only used for sequence/polymer results (for example after a sequence search)
 
 
 
 
Chemical Components
 

Integration with Binding Affinity Data from BindingDB

Binding Affinity Data on Structure Summary Page

Binding affinity data from BindingDB (http://www.bindingdb.org) have been integrated with the RCSB PDB website. The Ligand Chemical Component section of the Structure Summary page lists binding affinity data from BindingDB if they are available. These include binding constants IC50, EC50, Ki, and thermodynamic data Kd, ΔG, ΔH, -TΔS. If multiple experimental values are available for a measurement type, a range is listed. The data are linked to BindingDB details pages.

Show Ritonavir - HIV-1 Protease Complex (1HXW)

 
 
 

A new Advanced Search option finds PDB entries of protein-ligand complexes with associated binding affinity data. IC50, EC50, Ki, and Kd must be specified in nM, and thermodynamic data ΔG, ΔH, -TΔS in kJ/mol.

 
 
 

New Ligand Summary Page Layout

The ligand summary page has been redesigned with this release to use the widget framework used on the structure summary pages. It has also been redesigned to look and work more like the structure summary page. To increase the performance of the site, Jmol is no longer loaded by default.

Related Entries

The new Ligand Summary Page contains a 'Related Entries' widget. In the widget we provide links to all entries where the ligand appears as a polymeric residue, a free ligand, or both.

View Example (Ligand Summary Page for Biotin)

 

Related Ligands

We also include a widget that provides links to other related ligands. The search links available are for similar ligands, stereoisomers and a link to the chemical structure search using the ligand as the search basis.

View Example (Ligand Summary Page for HEM)

 
 
 
 
Comparison Tool Improvements
 

Standalone version supports alignment of SCOP domains

The latest version of the Comparison Tool (standalone application) now supports the alignment of SCOP domains. If the users enters the name of a SCOP domain (e.g. d4hhba_ ), the tool will automatically extract the domain boundaries from the SCOP classification. SCOP classification files will be cached in the same directory as is used for caching PDB files. A second improvement is that local database searches now can be run in multi-threaded mode, if the code is run on a multi-core computer.
 
 
Tabular Report Improvements
 

Ligand Summary Report on Ligand Hits Tab

The Ligand Hits tab now supports the generation of tabular reports, similar to the Structure Hits tab. A Ligand Results Table is available with Ligand ID, image, formula, molecular weight, name, SMILES string, and PDB IDs that contain the ligand.

 

All the tabular report features are available for Ligand Summary Report, such as the sorting, filtering, exporting to other report formats, and column customizing. The image on the left shows a Ligand Results Table filtered by the word "acid" in the ligand name column. Note, more advanced filter options are available by clicking on the "Filter Results" link on the bottom of the table.

 
 
 

New Fields for Customizable Table

Based on user requests the following fields have been added to the customizable tables:

  • Structure Summary
    • Macromolecule Type
    • Structure Molecular Weight
  • Sequence
    • Entity Macromolecule Type
    • Sequence and Secondary Structure
  • Ligands
    • Ligand Molecular Weight
  • Data Collection Details
    • Collection Date
    • Collection Temperature
  • Refinement Details
    • R Observed
    • R Work
    • R Free
    • R All
  • Primary Citation
    • DOI
    • MeSH Terms
    • Abstract Text Short

The new fields are well formatted in both tabular report and exported Excel spreadsheet. Shown is an example report that includes the Secondary Structure field.

 
 
 
 
General Site Improvements
 

Top-Bar Search

PubMed Abstract Search

PubMed Abstract text search has been added to the top-bar search. This search will match PDB structures that use the specified text term(s) in the PubMed abstract of their primary citation.

 
 
 

Browse Database

Auto-Complete

All hierarchical browsers for ontologies and classifications that are available both from the "Browse database" link on the left-side menu of the website, as well as, from the relevant advanced search options, support auto-complete functionality. This improves the usability of the interface since users may see directly, as they type, matching tree nodes, and by picking them, they will be transfered directly on the selected node in the hierarchy. Then, and after making sure that this is the location in the hierarchy that they are interested in, they can click on the selected node and get the associated structures.

The hierarchies that support auto-complete are:

The following Browsers (Browse Database tabs and Advanced Search options) support auto-complete: Biological Process (GO), Cellular Component (GO), Molecular Function (GO), Enzyme Classification, Source Organism (NCBI), Genome Location (NCBI), Medical Subject Headings (MeSH), SCOP, CATH.

 
 
 

Source Organism Browser with Common Names

Organism common names are now included in the organism naming information of the Source Organism taxonomy tree (NCBI) and they will be matched with searching and auto-complete.

This can be helpful in order to match against organisms and classes of organisms, without the need to remember (or consult another site for) the scientific name.

For example just typing "plants" on the Organism tree (Browse Database on the left side menu -> Source Organism tab) will reveal 4 matching taxonomy nodes with the word "plants" in the common name, and after examining them in the tree, we can select the higher level node "Viridiplantae (green plants)" to retrieve all the (about 2500) plant macromolecules in the PDB.

 
 
 

Sequence Search

PSI-BLAST Search to Find Distantly Related Proteins

The advanced sequence search now supports PSI-BLAST (Position-Specific Iterated BLAST), in addition to BLAST and FASTA searches. PSI-BLAST creates a profile (position-specific scoring matrix) derived from the multiple sequence alignment from an initial BLAST search and uses the profile to detect distant evolutionary relationships. PSI-BLAST applies an iterative search strategy during which the profile is refined. By default, 5 iterations are performed.

For more information see: http://www.ncbi.nlm.nih.gov/Education/BLASTinfo/psi1.html

 
 
 

Low Complexity Filter for Sequence Searches

A low complexity filter option has been added to the advanced sequence search. By default, low complexity regions in a sequence are filtered out to avoid spurious alignments. Low complexity regions are displayed as X in the query sequence.

For more information see: http://www.ncbi.nlm.nih.gov/Education/BLASTinfo/psi1.html

 
 
 

Sequence and Structure Visualization Improvement

Improved Domain Display on Sequence Tab

Domains are now displayed in a pseudo-3D rendering and domain annotations are displayed with a white background to increase readability.

 
 
 

New Jmol Display Options

The Jmol page has a new option to export images from Jmol in a variety of image formats.

 
 
 

Left Hand Menu

This release added a few new features to the left hand menu: MyPDB and Help widgets, a glossary of terms and a sitemap.

MyPDB Widget

The MyPDB left hand menu widget provides quick access to your MyPDB links while cleaning up the overall layout of the site. Once logged in, the widget displays links to your saved query manager, display settings and user account manager.

 
 
 

Help Widget

The help widget was added to give users better access to our help and tutorial content. For now it contains links to Advanced Search help, screencast tutorials and a new glossary of terms (see below). The Advanced Search help page is a list of all of our available advanced search options and explanations on how to use them. Video Tutorials is a section of screencasts made by the developers showing how to use various features of our site.

 
 
 

Glossary of Terms

A Glossary of Terms has been added to the website. This contains a listing of commonly used technical terms together with an explanation of each term. This is available from the Help box in the left hand menu.

Show Glossary of Terms

 
 
 

Sitemap

A link to a sitemap has been added to the left hand menu under the Home widget. This enables users to more quickly locate important pages of the website.

Show Sitemap

 
 
 

Website Performance Improvements

We have made various improvements to the architecture in order to increase the performance of our site. Many of our JavaScript libraries have been re-written, cleaned up and compressed. Our design layout is now much simpler, yet still very close to the last version and we have cleaned up our CSS. The positive upshot is reducing the number of calls to our servers the browser needs to make speeding up the site. Most of our content is also served up compressed as well, reducing the amount of data that needs to be transferred. We have also switched to using image sprites and inline image data allowing simpler DOM structure, less calls to the server and faster image load times. Our internal testing has showed that the site now renders 20-40% faster than the old production site and we expect the effect to be even greater overseas.

We used Yahoo.com's YSlow tool to help us rate the performance of the site as well as help us identify problems. According to their rating system, our old production site rated a 64 (D). This software release rates at between 89-93 (A) for various pages; a great improvement in performance.