|
|
Ligand-based design techniques use information
about one or several known actives (ligands) as a
basis for the design of lead compounds. It includes
applications to address critical ligand-based
design tasks, such as structure-activity
relationship modeling, pharmacophore hypothesis
generation, molecular alignment, and ADME
prediction. A ligand's shape is its most
significant property. There are no poorly fitting,
high-affinity ligands. Shape matters. If the three
dimensional coordinates of a ligand are known,
whether from experiment or rational analysis or can
be easily derived due to limited conformational
flexibility (or inspired guesswork), the problem of
finding a molecule of similar activity is
simplified. OpenEye approach is to define shape as
a rigorous metric property and search vast
conformational expansions for similar structures.
Others select points in space (pharmacophores, the
Cresset field points) that must be matched, or
generate partial descriptions of shapes from
fragments (Cramer's topomer method). Not
surprisingly, these all work to some degree or
other because molecular interactions occur in three
dimensions, not via strings.
|
ROCS- chemical
similarity analysis via rapid 3D molecular
shape searches
ROCS, a industry leading shape-matching
product, allows for fast comparisons of
shape, followed by simple matching of
chemical functionalities. ROCS shapes to
be defined either by molecule or by grid,
capable of distributing searches over
networks of computers. ROCS has never been
heard of failing. It is often so
successful at finding interesting leads,
physical screens have been cancelled. ROCS
results are easy to visualize and
comprehend and has been widely adopted by
medicinal chemists.
Seascape Scientific Partner:OpenEye
Scientific Software, New Mexico
|
|
EON- chemical
similarity analysis via comparison of
electrostatics overlay
Given an alignment of two molecules,
EON calculates and compares their
electrostatic fields. It can also perform
terminal rotor adjustments that do little
to change the shape match but that can
dramatically alter the electrostatic
profile. EON has been used to design
higher-affinity ligands (from a virtual
screen) and also as a method of
bio-isostere identification.
Seascape Scientific Partner:OpenEye
Scientific Software, New Mexico
|
|
BROOD-
bioisostere identification using shape,
chemistry and electrostatic
similarity
Bioisosteric replacement is a technique
used by medicinal chemists to modify the
physical or biological properties of lead
molecules without deleteriously affecting
their activity. Brood uses the shape,
chemistry and electrostatic similarity
technology it shares with ROCS and EON to
compare molecular fragments and identify
bioisosteres of a query fragment. Brood is
packaged with two pre-generated
multiconformer databases of molecular
fragment that are ready for searching. In
addition, it provides methods for
fragmenting molecular databases using a
default or user provided fragmentation
scheme. Brood can search more than 250,000
molecular fragments per hour, allowing
examination of common fragments in minutes
and query of larger more complex databases
in a few hours.
Seascape Scientific Partner:OpenEye
Scientific Software, New Mexico
|
|
QUACPAC-
quality charge states and charges for
small molecules and proteins
Calculating the correct partial charge
distribution for a compound is essential
if one wishes accurate electrostatic
fields or other electrostatic quantities.
One set of methods use electro-negativity
to apportion charge around a molecule.
Force-fields such as MMFF use this method
and OpenEye has implemented such. Better
physical agreement, however, can be found
by the Bayly method, AM1-BCC. Here, an
initial AM1 calculation is 'corrected' by
bond-types such that very high quality
(essentially ab initio level) can be
obtained very rapidly (~1/s).
Seascape Scientific Partner:OpenEye
Scientific Software, New Mexico
|
|
|
|
WABE-
electrostatics optimization of a lead
compound
If you know the shape is right, what
about better electrostatics? WABE is a
method and approach to maintain a similar
shape and to experiment with different
electrostatic potentials. Given a
potential distribution of an active
compound and a second molecule with the
same shape but different chemical graph,
WABE will generate analogs by a series of
isosteric replacements. WABE is capable of
generating anywhere from 10 to 10,000,000
potential leads, depending on the rules of
replacement, which are 'learnt' from
chemistries presented to it. Because the
shape remains the same the potential
comparison with the active molecule can be
made instantaneously (~100,000 per
second).
Seascape Scientific Partner:OpenEye
Scientific Software, New Mexico
|
|
VIDA II -
molecular visualization and data analysis
on very large datasets
Many of OpenEye tools either generate
or sift through vast quantities of 3D
ligand information, and as such VIDA II is
the best program available for large scale
visualization. VIDA II can handle small
lists of molecules or corporate
collections and can view them as
multi-pane 3D, 2D depictions, SMILES
strings, spreadsheet entries, html forms,
graphs and drill-down lists. It can view
them in the context of proteins or other
small molecules, as simple line models or
photo-realistic balls and sticks. With a
python core VIDA II is easily scriptable
and can be tailored to many end-uses.
Works, naturally, on many platforms.
Seascape Scientific Partner:OpenEye
Scientific Software, New Mexico
|
|
|
|
4D-QSAR -
Quantitative 3D Pharmacophore Models of
Biopotency
4D-QSAR analysis incorporates the
conformational, alignment, and
pharmacophore degrees of freedom in the
development of 3D-QSAR models. It is used
to create and screen against
3D-pharmacophore QSAR models and can be
used in receptor-independent or
receptor-dependent modes.
4D-QSAR can be used as follows:
- As a CoMFA pre-processor to provide
conformations and alignments
- In combination with CoMFA to
combine the field descriptors of CoMFA
with the GCOD descriptors of 4D-QSAR to
build a "best" model, or
- In addition CoMFA because it treats
multiple alignments, conformations and
embedded pharmacophores which are CoMFA
limitations.
The 4D-QSAR process:
- Model the 3D-pharmacophore
- View the active conformation
- Screen compounds using 4D-QSAR
model
Seascape Scientific Partner:
Chem21/Dr. Anton Hopfinger of Univ. of
Illinois.
|
|
ADMET/MI-QSAR
- Predict Performance Profiles of Drug
Candidates
ADMET/MI-QSAR permits the estimation of
a wide range of ADME and toxicity
endpoints based on the interaction of test
compounds with models of cellular
membranes and a set of unique property
descriptors.
ADMET properties are modeled based on
molecular interactions between test
compounds and cell membranes.
Quantitative structure-activity
relationship (QSAR) analysis relates the
magnitude of a property exhibited by a
molecule to its physicochemical and
structural parameters. Many ADMET features
are related to how the molecule interacts
with biological membranes. Some toxicity
endpoints, such as skin and eve
irritations, also depend on membrane
interactions. As part of the QSAR
model-building process, the ADMET/MI-QSAR
method takes these interactions into
account to treat the largest possible
range of ADMET biopharmaceutical and
toxicology problems.
Seascape Scientific Partner:
Chem21/Dr. Anton Hopfinger of Univ. of
Illinois
|
|
