Only a few short years ago, accessing
biomolecular structures from the
Brookhaven Protein Databank (PDB)
database involved purchasing a set of
magnetic tapes designed for a mainframe
computer. Structures and the critical information
about molecular function
they contain were out of reach to anyone
with a personal computer. But how
quickly things change. Nowadays, the
World Wide Web (WWW) has brought
forth several excellent services for freely
obtaining any file from the PDB structure
database (Box 1). Also, there are
two very good viewing programs, MAGE
(Refs 1, 2) and RASMOL (Ref. 3), both of
which have been made freely available
by their authors for use with IBM,
Macintosh and Unix systems. Kinemage
has a special niche as it allows extensive
annotation and description to be
carried along with the structure data,
making it a useful resource for teaching
4. (The journal Protein Science has a
WWW site containing a remarkable collection
of Kinemage format structure
files, each created individually by experts
and all freely available; see Box 1
for URL.)
Molecular Modeling Database
The Molecular Modeling Database
(MMDB) is the new structural division
of Entrez (Ref. 5), provided by the
National Center for Biotechnology Information
(NCB1). It holds all the structures
in the PDB database, but in a
different file format [specified in the
Abstract Syntax Notation 1 (ASN.1)
data description language6]. This format
allows files of structural data to be
readily compressed and exchanged between
modern computers. Our hope
is that with these transformed data,
structural scientists can start to design
and create software tools that allow
all of us to see different kinds of data,
such as structural superpositions and
non-atomic three-dimensional models
from electron microscopy 7, in a single
viewing environment.
The translation of PDB data files into
ASN.1 has involved the use of a
sophisticated PDB file parser*, which
was originally developed as part of an
ongoing research project into predicting
protein structure from sequence
data 8,9. This parser can 'sort' specific
pieces of data from a multitude of information,
and furthermore, can detect
and correct a variety of ambiguities that
arise in the PDB file format.
One common example of such an ambiguity
can be seen by viewing almost
any structure data file containing the
prosthetic groups NAD or FAD (e.g. 1LVL
Only a few short years ago, accessingbiomolecular structures from theBrookhaven Protein Databank (PDB)database involved purchasing a set ofmagnetic tapes designed for a mainframecomputer. Structures and the critical informationabout molecular functionthey contain were out of reach to anyonewith a personal computer. But howquickly things change. Nowadays, theWorld Wide Web (WWW) has broughtforth several excellent services for freelyobtaining any file from the PDB structuredatabase (Box 1). Also, there aretwo very good viewing programs, MAGE(Refs 1, 2) and RASMOL (Ref. 3), both ofwhich have been made freely availableby their authors for use with IBM,Macintosh and Unix systems. Kinemagehas a special niche as it allows extensiveannotation and description to becarried along with the structure data,making it a useful resource for teaching4. (The journal Protein Science has aWWW site containing a remarkable collectionof Kinemage format structurefiles, each created individually by expertsand all freely available; see Box 1for URL.)Molecular Modeling DatabaseThe Molecular Modeling Database(MMDB) is the new structural divisionof Entrez (Ref. 5), provided by theNational Center for Biotechnology Information(NCB1). It holds all the structuresin the PDB database, but in adifferent file format [specified in theAbstract Syntax Notation 1 (ASN.1)data description language6]. This formatallows files of structural data to bereadily compressed and exchanged betweenmodern computers. Our hopeis that with these transformed data,structural scientists can start to designand create software tools that allowall of us to see different kinds of data,such as structural superpositions andnon-atomic three-dimensional modelsfrom electron microscopy 7, in a singleviewing environment.The translation of PDB data files intoASN.1 has involved the use of asophisticated PDB file parser*, whichwas originally developed as part of anongoing research project into predictingprotein structure from sequencedata 8,9. This parser can 'sort' specificpieces of data from a multitude of information,and furthermore, can detectand correct a variety of ambiguities thatarise in the PDB file format.One common example of such an ambiguitycan be seen by viewing almostany structure data file containing theprosthetic groups NAD or FAD (e.g. 1LVL
การแปล กรุณารอสักครู่..
