The 2017.Jun release introduces the OEHermite class, which allows a molecule to be expanded into a Hermite representation. An initial suggested area of application is the representation of protein shape. The advantages of representing proteins by Hermites include:
Below are examples of low- and high-resolution Hermite representation images of dihydrofolate reductase (DHFR):
OpenEye will explore the potential of Hermite representations; we invite users to do likewise with this exciting new addition to the OpenEye toolkits. See the Shape from Hermite Representation chapter of the Shape TK documentation for more information.
Scalable Vector Graphics (SVG) image format is dominant in web applications due to its wide browser support and scalability. An SVG image contains a list of graphical objects that can be grouped, styled, and transformed together. This enables the generation of interactive images defined in the SVG XML elements or via scripting that accesses the SVG Document Object Model (DOM).
The new Preliminary API available in OEDepict TK provides low-level functionality that enables grouping of graphical elements in SVG images that can be manipulated externally by web developers. High-level API entrypoints are also available to generate self-contained SVG images that show or hide different elements of the image based on mouse events.
|Hover the mouse over any ligand atom to display its b-factor||Click on any residue circle to display ligand atoms it interacts with|
For more information see the Generating Interactive SVG images chapter of the OEDepict TK documentation.
OEGraphSim TK now provides improved fingerprint search capability, allowing users to perform molecule similarity searches on large databases up to 10x faster than previous versions.
Fingerprint methods are used for the computation of similarity scores. When searching large virtual databases, performance can be problematic. To address this issue, OEGraphsim TK now uses popcount, a hardware instruction available on modern CPUs, to provide significant performance improvements for computing similarity.
The new fingerprint search can be performed in two modes: in-memory and memory-mapped. The in-memory mode involves preloading all fingerprints into memory and performing the search. While this represents the fastest way to perform similarity searches once the fingerprints are loaded, searches are limited by memory availability. The memory-mapped mode has no load time penalty or memory limitation but the search itself takes more time.
|fingerprint generation performance||fingerprint search performance|
OpenEye continues to monitor and collect feedback from our customers on the need to support Python 2 wrappers. Our current plan is to phase out support for Python 2 on all platorms by the 2017.Oct release. As this is a substantial change for our customers, we are willing to help with migration. Please contact firstname.lastname@example.org for more details.
|Hover the mouse over any atom||Click on any atom|
|Hover the mouse over any bond||Click on the middle on any bond|
On OSX, the deprecated CGContextSelectFont and CGContextShowTextAtPoint Core Graphics functions have been replaced. These function were used to handle fonts when drawing text in PNG images. OEDepict TK currently supports the following font mapping on OSX for PNG images:
|OpenEye Font Family Type||Font Name|
The following new getter methods have been added to OEShapeDatabaseOptions for retrieving the number of starts for the relevant method:
OEShapeDatabaseOptions::GetNumStarts has also been overloaded for each alternative start option.
There is no need to alter existing scripts if the OEFastROCSOrientation::AsIs alternative starts feature is not being used since the default argument results in the correct execution path for all other use cases.
Generally, the constructor is not used in most scripts as the OEShapeDatabaseScore object is constructed internally when calling OEShapeDatabase::GetScores or OEShapeDatabase::GetSortedScores. Therefore, most users will not be affected by this change.
The default value for the additional argument satisfies all other use cases, so no action is required.
See OEShapeDatabaseScore::Constructors for more information.
Previously, GPU memory limits were being reached for routine searches when using OEFastROCSOrientation::InertialAtHeavyAtoms, causing searches to default to OEFastROCSOrientation::Inertial. The improved alternative starts algorithm handles any number of starts without reaching GPU memory limits.
Performance will be affected by using large numbers of alternative starts.
OEShapeDatabaseOptions::SetMaxOverlays now sets the number of overlay results to return per conformer. Previously, this was broken and was restricting the number of overlays to 1 per conformer regardless of its setting. OEShapeDatabaseOptions::SetMaxOverlays is now fixed and works correctly with OEShapeDatabaseOptions::SetMaxConfs, OEShapeDatabaseOptions::SetNumInertialStarts and alternative starts for both OEShapeDatabase::GetScores and OEShapeDatabase::GetSortedScores.
Tversky simulations no longer overestimate the score of caged systems or conformer pairs where one conformer is much larger than the other.
A tutorials page has been created that walks through features of FastROCS TK and gives detailed examples. The first tutorial gives step-by-step instructions on how to use alternative starts.
If you have a specific feature that you would like in a tutorial, please contact support at email@example.com with the subject header FastROCS TK Tutorial Request.
click on any residue
Example of interactive SVG image generated by OERenderContactMap (PDB: 4HWR)
Several problems in the OERenderActiveSite function have been fixed:
OEDrawPeptide has been fixed and now allows only amide and disulfide bonds when identifying standard amino acids. No alpha-carbon substitutions are allowed.
A new API has been added to perform rapid searching of fingerprints using the popcount method:
GraphSim TK currently only supports the popcount search method for fingerprints where the size is multiple of 256. This means that the OEFPType::MACCS166 fingerprint type is currently not supported. See the User-defined Fingerprint section.
The above-mentioned OESimScore class API change means that the following warning messages could be encountered when compiling old code with the new GraphSim TK:
implicit conversion loses floating point precision: 'double' to 'float'
for (OEIter<OESimScore> si = fpdb.GetScores(mol); si; ++si) float score = si->GetScore(); // GetScore() now returns double
implicit conversion loses integer precision : 'size_t' to 'unsigned int'
for (OEIter<OESimScore> si = fpdb.GetScores(mol); si; ++si) unsigned int molidx = si->GetIdx(); // GetIdx() now returns size_t
This is the first official release of OEFF TK. The underlying functionality was previously released in C++ as part of the OESzybki TK advanced interface. This toolkit is currently available in C++ and Python. An overview of its capabilities is provided below.
OEFF TK provides access to powerful low-level functionality for advanced users to develop molecular modeling applications. With OEFF TK, users also have the ability to create variations of molecular interaction classes for energy evaluation and optimization, adding to the built-in interactions, if desired. OESzybki TK complements OEFF TK as the high-level API for molecular modeling functionalities.
All molecular interactions are objective functions with additional API points for molecular definitions. A set of pure virtual classes that can be used to define new interactions are provided for both generic objective functions and molecule specific interaction functions. The following virtual base classes are available:
Force fields and components of force fields are defined as molecule objective functions and are derived from the above-specified base classes. The following force fields and components are available:
A powerful set of optimizers that can be used to optimize objective functions is now available:
Adaptors that can be used to optimize molecule objective functions in transformed coordinate space are now available:
Two new intermolecular force fields, OEMMFFAmber and OEMMFFIEFF, have been added to the force field suites for molecular interactions. Both force fields use OEMMFF for intermolecular interactions and more accurate components for intermolecular interactions.
A new class, OESheffieldOptions, has been added that sets all options of the OESheffield object during its construction. All OESheffield methods with similar names as their OESheffieldOptions counterparts are now deprecated.
The following table shows the older, deprecated functions and their new option replacements:
|OESheff::OESheffield::GetParameters||OESheffieldOptions::GetA OESheffieldOptions::GetB OESheffieldOptions::GetSoluteDielectricConstant OESheffieldOptions::GetSolventDielectricConstant|