OpenEye toolkits are increasingly being used in web services that require protection from malicious users. The most obvious attack vector against the OpenEye toolkits is file format parsing since scientific file formats are complex and often underdefined.
OpenEye has closed a significant number of vulnerabilities related to the parsing of supported file formats: molecules, grids, surfaces, receptors, etc. In addition, state-of-the-art testing and detection techniques are now continuously run on the OpenEye code base to ensure a high level of security going forward.
Security is a task requiring continuous vigilance and OpenEye will continue to make this effort a high priority.
The utility of FastROCS has always been hampered by the amount of physical memory available in GPU machines. The computational chemistry community is continuously pushing the boundaries of FastROCS by searching larger and larger virtual libraries of compounds. In addition, FastROCS servers are becoming an increasingly important component of many infrastructures and must be able to be rapidly brought up and down on demand. Both these problems come down to the fact that loading a database into memory is often between 10 and 20 times slower than performing an actual FastROCS search on that database.
This problem has been addressed in this release by dramatically increasing the performance when loading a dataset into memory for eventual FastROCS processing. OEPrepareFastROCSMol generates additional information to be written to OEB that is used to rapidly read molecules into memory.
Using OEPrepareFastROCSMol on a molecule can result in load times that are nearly 10 times faster than previous uncached OEB file loads. Depending on the ratio of CPU and GPU compute power and conformers per molecule, this can mean that loading a dataset can now be just as fast as the FastROCS search itself.
OpenEye is pleased to announce the first official release of OEMedChem TK. This medicinal chemistry functionality was previously only available in beta form.
OEMedChem TK provides the ability to index a set of input structures and identify matched molecular pairs. Matched molecular pair analysis is becoming recognized as a powerful tool for the extraction of the effects of chemical changes on a property or properties of interest in large data sets.
Additional OEMedChem TK capabilities include fragmentation and perception, similarity measures, belief theory, and molecular complexity measures. For the full list of capabilities, see the OEMedChem functionalities section.
If you would like to utilize the functionalities of OEMedChem TK, please contact firstname.lastname@example.org for a license.
The Python 2 distributions of OpenEye Python Toolkits can now handle both ascii and unicode Python strings for all APIs that accept string arguments. For example, the OESmilesToMol function can now parse the following SMILES strings without throwing a ‘TypeError’ exception:
smiles_str = "[N+](=O)([O-])([O-]) nitrát" smiles_unicode = u"[N+](=O)([O-])([O-]) nitrát"
Previously, the unicode strings would work in Python 3, but not in Python 2, making it difficult to write code that was compatible with both major versions of Python at the same time. This major change should make porting to Python 3 much easier for OpenEye toolkit users.