RCSB PDB Newsletter
Number 33 -- April 2007
Published quarterly by the
Research Collaboratory for Structural Bioinformatics Protein Data Bank
Weekly RCSB PDB news is published at www.pdb.org
To change your subscription options, please visit
lists.sdsc.edu/mailman/listinfo.cgi/rcsb-news
-----------------------------------------
TABLE OF CONTENTS
Message from the RCSB PDB
Data Deposition and Processing
Restarting ADIT Depositions
Weekly Deadlines for Release/Modify Entry Requests
Deposition Statistics
Depositing and Releasing Experimental Data
Data Query, Reporting, and Access
New Website Features
RCSB PDB Focus: Saving Protein Workshop "States" for Future
Visualization Sessions
Website Statistics
Outreach and Education
Citing Structures in the PDB: IDs, Citations, and DOIs
New Information and Statistics Available at BioSync
Making Virus Models with Middle School Students
Molecules of the Quarter
PDB Education Corner: New Jersey Science Olympiad
PDB Community Focus: Angela Gronenborn, University of Pittsburgh
Statement of Support, Partners, Leadership Team
Snapshot
--------------------------------------------
MESSAGE FROM THE RCSB PDB
The RCSB PDB fosters communication with users to support access to the
information contained within the PDB archive. At the same time, our
outreach efforts are focused on soliciting input from the user
community to help improve RCSB PDB services. Frequently, this dialog
happens at workshops and professional society meetings.
Recent activity has included:
* The Keystone Symposia "Frontiers of NMR in Molecular Biology"
(January 6-11 in Snowbird, Utah). A workshop entitled The Future of
Publicly-Accessible Databases for NMR Spectroscopy was held to
discuss issues surrounding deposition and data representation in the
PDB and BMRB. These issues were further examined at a meeting of
the PDB-BMRB joint NMR Task Force held after the workshop.
* Annotators Jasmine Young and Monica Sekharan exhibited at the 51st
Annual Meeting of the Biophysical Society (March 3-7 in Baltimore,
Maryland).
* The RCSB PDB also exhibited at the Celebration of Teaching &
Learning, an education-related professional development conference
for teachers, administrators, and others (March 23-24 in New York
City).
We hope to see many of you at our upcoming meetings, some of which are
highlighted below:
* The Experimental Biology Annual Meeting for several professional
societies, including the American Society for Biochemistry and
Molecular Biology (April 28 - May 2 in Washington, DC).
* The 5th Annual International Conference on Intelligent Systems for
Molecular Biology (ISMB) & 6th European Conference on Computational
Biology (July 19-25 in Vienna, Austria).
* The meeting of the American Crystallographic Association (ACA; July
21-26 in Salt Lake City, UT).
* The 21st Annual Symposium of The Protein Society (July 21-25 in
Boston, MA).
* The 9th International Conference on Biology and Synchrotron
Radiation (August 13-17 in Manchester, United Kingdom).
* The 234th American Chemical Society National Meeting (August 19-23
in Boston, MA).
The RCSB PDB Poster Prize will also be awarded at ISMB, ACA, and the
the European Crystallographic Meeting.
--------------------------------------------
DATA DEPOSITION AND PROCESSING
Restarting ADIT Depositions
A structure can be deposited in more than one Internet session by
using ADIT's "Session Restart ID" feature. This identifier appears in
red in the center of the browser window when ADIT's "deposit" step is
first started. It is also seen in the title of the browser throughout
the deposition session. The case-sensitive restart ID should be
entered in the space provided on the ADIT home page to return to the
undeposited entry. Any data entered in a category are stored every
time the user selects the SAVE button. All entered data associated
with a particular entry can be accessed using the restart ID until the
"DEPOSIT NOW" button is selected, for up to six months after the
session has been last updated. ADIT is available at the RCSB PDB and
PDBj. ADIT-NMR can be used to deposit data to the PDB and BMRB at the
same time. ADIT tutorials (in English and Japanese) and example "in
progress" deposition sessions are accessible from deposit.pdb.org.
Weekly Deadlines for Release/Modify Entry Requests
PDB entries are processed by three members of the wwPDB (RCSB PDB,
MSD-EBI, and PDBj). Each week, all files scheduled for release or
modification are checked and validated one final time. Authors may be
contacted to resolve any issues that may arise while preparing the
entries for release. When the release of a structure on hold for
publication (HPUB) is requested, the wwPDB routinely confirms the
primary citation. If this is not accomplished within that release
cycle, the entry may be scheduled for release in a later update.
To be included in the next weekly update, any required author
correspondence should be sent by 15:00 (local time) on Thursdays to
the appropriate wwPDB member: RCSB PDB (help@deposit.rcsb.org);
MSD-EBI (pdbhelp@ebi.ac.uk); PDBj (adit@adit.protein.osaka-u.ac.jp)
All entries due for release are transferred to the RCSB PDB for final
packaging into the master PDB ftp archive. These files are then
released by 4:00 ET each Wednesday. Requests received after these
cutoff times will be processed during the next update cycle.
Deposition Statistics
In the first quarter of 2007, 2319 experimentally-determined
structures were deposited to the PDB archive. Of these structures,
62.7% were deposited with a release status of HPUB, 22.3% with a
release status of hold until a particular date, and 15.0% were
released as soon as annotation of the entry was complete. 80.8% of
these entries were determined by X-ray crystallographic methods; 18.8%
were determined by NMR methods. 80.3% of these depositions were
deposited with experimental data. 92.8% of the crystal structures
were deposited with structure factors; 27.7% of NMR structures were
deposited with restraints.
Depositing and Releasing Experimental Data
The RCSB PDB strongly encourages depositors to follow the guidelines
regarding the submission and release of coordinate and experimental
data that have been set by the International Union of Crystallography,
the National Institutes of Health, and the journals. Deposition of
experimental data (structure factor and/or NMR constraint files) is
required by many journals, including Acta Crystallographica,
Biochemistry, Cell, Nature, and Science. These files can be uploaded
during the ADIT deposition process. Depending upon the hold status
selected by the depositor, data release can occur when a depositor
gives approval, the hold date has expired, or the journal article has
been published. There is a one-year limit on the length of a hold
period, including HPUBs. If the citation for a structure is not
published within the one-year period, depositors will be given the
option to either release or withdraw the deposition. Detailed
deposition and release information is at deposit.pdb.org.
--------------------------------------------
DATA QUERY, REPORTING, AND ACCESS
New Website Features
Many enhancements were made for data query and reporting this quarter.
* Improved Access to Ligand Data
The PDB chemical component dictionary (formerly the HET dictionary)
has been remediated to better describe the components that interact
with macromolecular structures. This new dictionary has been
incorporated with the RCSB PDB database.
A new search results option is a tab called "Ligand Hits". This page
lists the ligands known to interact with the structures that match the
query.
For example, a search for "protein kinase" returns 2045 structures and
676 ligands. From the "Ligand Hits" page, users can find all of the
structures that contain that ligand or access information from the
"Ligand Summary" page. This page offers summary information, downloads
(definitions and coordinates), and interactive and static views.
* Ligand Explorer Tool for Viewing Protein-Ligand Interactions
Ligand Explorer is a Java-based program accessible from each Structure
Summary page. Features include the ability to highlight ligand
interactions based on conventional and user-defined thresholds, and a
"contact map" that gives users the ability to see the details of each
interaction.
* Access to Single Nucleotide Polymorphism (SNP), Pfam, and More
SNP information is now accessible from Structure Summary pages. Over
4000 PDB structures are linked to SNP information from the SNP
database. This information is accessible from each entry's "Biology
and Chemistry Report" tab. The Pfam database contains multiple
alignments of protein domains. With each release of the Pfam data,
files mapping Pfam domains to PDB structures are made available on the
Pfam FTP site. This mapping is loaded into our database so that Pfam
domain information for a protein structure is displayed on an entry's
Structure Summary page and Biology and Chemistry Report, when
available.
The "External Links" option provides further information about the
structure under study, such as biochemical pathway information,
stereochemistry and ligand binding data. When looking at an entry's
Structure Summary page, the external links page is accessible from the
left-hand menu.
SNP: www.ncbi.nlm.nih.gov/projects/SNP
Pfam: www.sanger.ac.uk/Software/Pfam
* New Advanced Search Options
Simple searches of the RCSB PDB website can be performed using the
keyword box at the top of each page. The "Advanced Search" feature
makes more specific and complex searches possible. New options can be
used to search the PDB using keywords, phrases or a series of
keywords:
- Advanced Keyword Search: This option can be used to search for
keywords in the full text or by the author name. If you enter a
phrase, you must place it in quotes otherwise it will be interpreted
as a series of keywords. Advanced keyword search supports the lucene
syntax for sophisticated string searching.
- Medical Subject Headings (MeSH): Searches for structures associated
with particular MeSH terms from the National Library of Medicine
(NLM). This option launches the MeSH Browser, which lets users either
browse through the MeSH hierarchical tree or search the tree with
keywords.
- Author Assigned: Looks for structures based upon keywords used by
the depositor.
- PubMed: Searches PubMed titles and abstracts for an entry's primary
citation (if it exists).
* New Help Features
A new set of Flash Tutorials, modeled on the popular guides on how to
use the RCSB PDB overall site and the Advanced Search, are available.
These include tutorials for the MeSH Browser, Protein Workshop, KiNG,
Jmol, and general navigation. They are accessible from the left-hand
menu under "Site Tutorials".
Quick Tips offer hints and quick links to exploring the RCSB PDB
website. To view them, click on the "Show Quick Tips" in the
left-hand menu. Clicking on the arrow button will scroll through
these hints, and clicking on the "X" will close the box.
Time-stamped Copies of PDB Archive Available via FTP
A time-stamped snapshot of the PDB archive as of January 2, 2007 has
been added alongside time-stamped copies of the archive from January
2006 and 2005 at ftp://snapshots.rcsb.org. It is hoped that these
snapshots will provide readily identifiable data sets for research on
the PDB archive.
The directory 20070102 includes the 40,933 experimentally-determined
coordinate files that were current (i.e., not obsolete) as of January
2, 2007. Coordinate data are provided in PDB, mmCIF, and XML
formats. The date and time stamp of each file indicates the last time
the file was modified.
Scripts can be used to automatically download data:
* ftp://snapshots.rcsb.org/rsyncSnapshots.sh Makes a local copy of an
annual snapshot or sections of the snapshot. Downloading the entire
archive can be lengthy (more than 18 hours), but the time required
to download data in a single format should be much less. Depending
upon network speed, our tests show that all of the coordinate files
in PDB format from a snapshot can be down loaded in about 2 1/2
hours.
* ftp://ftp.rcsb.org/pub/pdb/software/rsyncPDB.sh
Copies the current contents of the entire archive.
* ftp://ftp.rcsb.org/pub/pdb/software/getPdbStructures.pl
Copies portions of the current archive.
* ftp://ftp.rcsb.org/pub/pdb/software/getPdbUpdate.pl
Copies the data from the weekly updates.
RCSB PDB Focus: Saving Protein Workshop "States" for Future
Visualization Sessions
Protein Workshop is a molecular viewer accessible from every PDB
entry's Structure Summary page. Its simple interface lets users
quickly and easily select structural elements and change the coloring,
labeling, and representation style (ribbons, cylinders, and
more). Users can also color specific structural features such as
conformation type and hydrophobicity.
Protein Workshop is an excellent tool for generating high-resolution
images in JPG, BMP, TIFF, WBMP, and PNG formats. A tutorial for using
Protein Workshop and creating these images is available.
Protein Workshop offers a way to save the "state" of a session. Users
can rotate and zoom a structure to a particular orientation and then
capture this view for later use. To save a state, enter a title next
to the "Capture current viewer state" from the Options menu, and then
select the adjacent button. The name of this state will be listed in
the box below. The view of the molecule can then be changed around,
but users can always go back to saved states by clicking on the
state's name.
These states can be saved in a XML file for later use by selecting the
state and clicking the "Export selected state" button. States can be
restored from a file by clicking the "Import state" button.
This tool uses the Molecular Biology Toolkit (mbt) and JOGL
technology, and requires no installation other than the most recent
version of Java.
Website Statistics
Access statistics for the first quarter of 2007 are given below for
the RCSB PDB website at www.pdb.org.
Month Unique Number of Bandwidth
Visitors Visits
Jan 2007.....124450.....299114.......503.06 GB
Feb 2007.....118164.....282552.......449.78 GB
Mar 2007.....125905.....298491.......472.08 GB
--------------------------------------------
OUTREACH AND EDUCATION
Citing Structures in the PDB: IDs, Citations, and DOIs
The contents of the PDB are in the public domain. Structures can be
cited using their PDB ID and the published citation related to the
structure.
* Structures may also be referenced using their Document Object
Identifier (DOI). The DOIs for PDB structures all have the same format
- 10.2210/pdbXXXX/pdb - where XXXX should be replaced with the desired
PDB ID. For example, the DOI for PDB entry 4HHB is
10.2210/pdb4hhb/pdb. This DOI can then be used as part of a URL to
obtain the entry's compressed data file in PDB format
(http://dx.doi.org/10.2210/pdb4hhb/pdb), or can be entered in a DOI
resolver (such as http://www.crossref.org) to automatically link to
pdb4hhb.ent.Z in the PDB ftp archive (ftp://ftp.rcsb.org).
* The journal reference for the RCSB PDB is: H.M.Berman, J.Westbrook,
Z.Feng, G.Gilliland, T.N.Bhat, H.Weissig, I.N.Shindyalov,
P.E.Bourne. (2000) The Protein Data Bank, Nucleic Acids Research
28:235-242.
* The journal reference for the wwPDB is: H.M. Berman, K. Henrick,
H. Nakamura. (2003) Announcing the worldwide Protein Data
Bank. Nature Structural Biology 10:980.
Detailed information for citing the use of data, structures, and
images is available from the RCSB PDB home page under "General
Information".
New Information and Statistics Available at BioSync
The BioSync website now contains updated beamline descriptions for
operational US synchrotron beamlines as well as some basic information
for almost all operational international beamlines.
PDB deposition statistics, grouped by site and beamline, can be found
at biosync.rcsb.org. Galleries of structures, also grouped by site and
beamline, are cross-linked to Structure Summary pages in the RCSB PDB.
Tables of primary citations and some general information (phasing
software, resolution, R-factors, etc.) are also provided. Most
recently, similar tables and galleries have been added for structural
genomics structures solved from synchrotron data.
Updates to beamline descriptions from local personnel, general
comments, and suggestions should be sent to BioSync@deposit.rcsb.org.
Making Virus Models with Middle School Students
Annotators helped middle school students in New Jersey build 3D models
of virus structures as part of Princeton University's Science and
Engineering Expo (March 22). For a copy of the template used to create
paper virus models, please email info@rcsb.org.
Molecules of the Quarter
The Molecule of the Month series explores the function and
significance of selected biological macromolecules for a general
audience. The molecules featured this quarter were importins,
exosomes, and zinc fingers. The complete features are accessible from
www.pdb.org.
--------------------------------------------
PDB EDUCATION CORNER: NEW JERSEY SCIENCE OLYMPIAD
Science Olympiad tournaments, which take place across the country,
consist of a series of individual and team events that students
prepare for during the year. During this competition, teams
demonstrate their diverse skills and knowledge in many different
events. In Forensics, teams identify polymers, solids, and fibers at a
crime scene, while in Write It, Do It, students compose a description
of a structure that will be the only guide used by their other team
members to recreate that structure (sight unseen) with raw materials.
High school teams at the New Jersey Science Olympiad (NJSO)
demonstrated their understanding of structure and function in the 2007
Protein Modeling trial events that were sponsored by the RCSB PDB.
In this event, students identify key elements of a structure and
demonstrate their knowledge of the protein by creating a
three-dimensional model using Mini-Toobers, computer visualization
tools, and RCSB PDB resources. The model is accompanied by a brief
abstract that highlights the features shown in their model and
discusses what the protein does. At the competition, teams also answer
multiple choice and short answer-questions focusing on its structure
and function.
As one team described, the students enjoyed "replicating protein
molecules that are found in the body into real-life toober models".
The entries are judged by the RCSB PDB annotators using a model built
directly from the structure's PDB file and a predetermined rubric that
awards points for accurate depictions of the protein's features. For
example, judges look to see if the N- and C- terminus are labeled
properly and carefully consider the helices of the model. They also
consider if the main functional and structural features of the protein
are described in the written abstract. The written exam asks questions
based upon the entry's Structure Summary page, the Molecule of the
Month entry, and beyond. In 2007, teams built an insulin structure
(PDB ID 4hiu) for the regional competitions held in January, and a
section of a major histocompatibility complex (MHC) structure (PDB ID
1hsa) for the state competition in March. The hand-built models were
really impressive, and the written abstracts and exams exhibited that
many teams were quite scientifically literate.
At the Central New Jersey regional, East Brunswick High School (First
Place and the 2006 State Champions in this event), West
Windsor-Plainsboro South High School (Second), and West
Windsor-Plainsboro North High School (Third) created very strong
models. At the Northern New Jersey regional, Bergen County Academy
(First Place), Westfield High School (Second), and New Providence High
School (Third) exhibited very strong skills. At the state finals,
students from all over the garden state competed. The highest ranked
teams were Princeton High School (First Place), Montgomery High School
(Second), and The Lawrenceville School (Third).
The Science Olympiad is an international nonprofit organization
devoted to improving the quality of science education, increasing
student interest in science and providing recognition for outstanding
achievement in science education by both students and teachers. The
2007 NJSO (www.njscienceolympiad.org) was presented by the New Jersey
Science Teachers Association and the New Jersey Science Education
Leadership Association. Special thanks to the Center for BioMolecular
Modeling at the Milwaukee School of Engineering (www.rpc.msoe.edu/cbm)
for the design of this event. Kits similar to those provided for this
event may be purchased from www.3dmoleculardesigns.com. Questions
about the NJSO Protein Modeling trial event should be sent to
buildmodels@deposit.rcsb.org. A website with information and
resources for participating in the protein modeling event can be found
at education.pdb.org/olympiad.
For an excerpt of one of the abstracts, please see
www.rcsb.org/pdb/general_information/news_publications/newsletters/2007q1
--------------------------------------------
PDB COMMUNITY FOCUS: Angela Gronenborn, University of Pittsburgh
Angela Gronenborn, Ph. D. is one of the country's leading structural
biologists and an internationally renowned specialist in the
application of nuclear magnetic resonance (NMR) spectroscopy for
investigating structure, dynamics and folding of biological
macromolecules. She joined the faculty of the University of Pittsburgh
as a Professor in the School of Medicine in 2004. In 2005, the
Department of Structural Biology was established with Prof. Gronenborn
holding the Rosalind Franklin Professorship and Chair. The department
is located in the new Biomedical Science Tower, housing state of the
art equipment devoted to NMR spectroscopy, X-ray crystallography, and
cryo-electron microscopy.
Prior to her move to Pittsburgh, Prof. Gronenborn was a member of the
Senior Biomedical Research Service and Chief of the Structural Biology
in the National Institute of Diabetes and Digestive and Kidney
Diseases at the National Institutes of Health (NIH). She received both
her undergraduate and Ph.D. degrees from the University of Cologne,
Germany. After post-doctoral training she joined the Scientific Staff
in the Divisions of Molecular Pharmacology and Physical Biochemistry
at the National Institute for Medical Research, Mill Hill, London. In
1984, she moved to the Max-Planck Institute in Munich as head of the
Biological NMR Group, and in 1988 to the NIH. Prof. Gronenborn's
research harnesses the power of NMR in two major areas: understanding
biochemical mechanisms and the structural basis of cellular regulation
as well as HIV pathogenesis. She has authored more than 350
publications, including structural studies on interleukins,
chemokines, the tumor suppressor protein p53, various transcription
factors and enzymes, and a number of HIV-encoded proteins including
integrase and protease. She also is noted for her contributions to
advancing technology on how best to apply NMR to elucidate important
problems in the biosciences.
Q: How would you compare X-ray and NMR methods for determining structure?
A: I truly believe both methods are complementary. Each provides a
model for the 3D structure of a molecule and as such, each presents a
picture of the spatial architecture, one in the solid state and the
other in solution. Naturally, the environment and conditions in which
the structural studies are conducted will influence the outcome to a
certain degree. pH, temperature, and ionic strength are rarely
identical if both methods have yielded structures, and details may
vary accordingly. For example, sidechain orientations may differ
depending on the protonation state, and loop regions may get "locked
in" in the crystalline state. In addition, since a much larger degree
of order is required for crystals to form, the oligomerization state
may be different in solution and the crystal. Indeed, there are
numerous examples of proteins for which dimers and higher oligomers
are observed by X-ray crystallography, but the solution NMR structures
are monomeric.
In terms of methodological maturity, it is evident that X-ray
crystallography is 25 years ahead of NMR as a structural method, thus
it is a robust method. This is reflected in the significantly larger
numbers of X-ray structures in the PDB compared to NMR structures. If
one looks at the growth rate, however, I believe NMR follows exactly
the trend that was seen 25 years ago in the crystallographic
field. Structural NMR is still evolving, with novel and advanced
approaches being introduced all the time. A case in point was the
introduction of R(esidual) D(ipolar) C(coupling)-based methodologies
that led to better defined structures and allows for unambiguous
positioning of relative structural elements.
Q: What aspects of structural biology are more accessible by NMR than
X-ray methods?
A: As we all know, the rate-limiting step in X-ray crystallography
frequently is the time it takes to obtain well-diffracting single
crystals--NMR solution structural work is not hampered by this
requirement. Crystallization may be prevented if, for instance, a
protein is very flexible or contains mobile regions, but NMR can
investigate such "floppy" proteins. Examples of this type are folding
intermediates or partially folded proteins, for which NMR is probably
the only method that allows one to carry out structural
characterizations (see for example ref.1).
In addition, structures of weakly interacting systems are another area
where NMR excels. Tight binding is often required for complexes to be
amenable to crystallization, and exchanging systems present major
challenges (sometimes overcome by cross-linking the components). NMR
can deal with exchanging systems and structures of "weak" complexes
can be determined (see for example ref.2). This property of NMR was
exploited early on in studies of protein-ligand complexes and the
transferred NOE methodology has been widely used in pharmaceutical
applications.
Q: What were the most exciting projects in which you have been
involved?
A: There have been numerous exciting projects all along the way--and
I still can get thrilled about seeing a new structure for the first
time or coming up with some crazy idea.
One exhilarating period that comes to mind was the late eighties/early
nineties when we were all in the bowels of Building 2 at the
NIH. Marius Clore and myself had just moved from the
Max-Planck. Together with Ad Bax, who already was working there, and a
combined group of congenial post-docs, we developed and implemented 3-
and 4D NMR and its application for protein structure determination.
Also, my work on cyanovirin (CVN)--starting with the initial
structure of a protein whose sequence had no relatives in any database
via dissecting its folding and domain-swapping, to carbohydrate
binding and the structural basis of its anti-HIV activity has kept me
captivated for years. Indeed, it inspired me to embark on a fishing
expedition -- for its gene -- which in turn has now led to the discovery
of CVN homologs in truffles and plants, whose structures we are
currently working on.
For the rest of the interview, which discusses the BMRB, the PDB,
changes in NMR techniques over the years, and the challenges of
establishing a structural biology department at the University of
Pittsburgh, please see
www.rcsb.org/pdb/general_information/news_publications/newsletters/2007q1
----------------------------------------
STATEMENT OF SUPPORT
The RCSB PDB is supported by funds from the National Science
Foundation, the National Institute of General Medical Sciences, the
Office of Science, Department of Energy, the National Library of
Medicine, the National Cancer Institute, the National Center for
Research Resources, the National Institute of Biomedical Imaging and
Bioengineering, the National Institute of Neurological Disorders and
Stroke, and the National Institute of Diabetes & Digestive & Kidney
Diseases.
The RCSB PDB is managed by two partner sites of the Research
Collaboratory for Structural Bioinformatics:
RUTGERS
Rutgers, The State University of New Jersey
Department of Chemistry and Chemical Biology
610 Taylor Road
Piscataway, NJ 08854-8087
SDSC/Skaggs/UCSD
San Diego Supercomputer Center and the Skaggs School of Pharmacy and
Pharmaceutical Sciences
University of California, San Diego
9500 Gilman Drive
La Jolla, CA 92093-0537
RCSB PDB LEADERSHIP TEAM
Dr. Helen M. Berman - Director
Rutgers University
berman@rcsb.rutgers.edu
Dr. Philip E. Bourne - Co-Director
SDSC/Skaggs/UCSD
bourne@sdsc.edu
A list of current RCSB PDB Team Members is available from the website.
The RCSB PDB is a member of the Worldwide PDB (www.wwpdb.org)
--------------------------------------------
SNAPSHOT
April 1, 2007
42474 released atomic coordinate entries
* Molecule Type
39000 proteins, peptides, and viruses
1713 nucleic acids
1726 protein/nucleic acid complexes
35 other
* Experimental Technique
36086 X-ray
6159 NMR
144 electron microscopy
85 other
25371 structure factor files
3377 NMR restraint files