Software

Our flagship product GRACE has been developed with and for our customers for over twenty years. From the early days up until today, we have been tightly connected to the experimental departments of some large pharmaceutical companies, embracing their problems as ours.

Our mission is to rush ahead to enable our customers to reliably predict the solid-state properties of their compounds. Since the beginning, we had to run ahead of the crowd: first, competing with numerous academic groups already established in the field of crystal structure prediction and then with other companies, being much larger, benefiting from substantial amounts of venture capital or both. The fact that we never lost our lead definitely proves one thing: we are innovating rapidly and cost-effectively.​

We happily integrate third-party components, along with accepting to pay licence fees if that is the best choice for us and our customers. However, for most problems we are facing, we need to go our own way. Our software solutions are robust workflows with a convenient command line interface based on template files. In addition, our workflows run stable for weeks or months on hundreds to thousands of cores, utilizing in-house or cloud High Performance Computing (HPC) Linux clusters. As such, wasting development resources on Graphical User Interfaces (GUIs) or vertical control of all software components is not on our agenda.

Large pharmaceutical companies, who want to apply our technology to their entire small molecule portfolio, often start to use GRACE themselves in-house or on the cloud at some point. GRACE is running on many different LINUX clusters, featuring different types of schedulers (SLURM, SGE, etc). User training has always been straightforward, hardly ever requiring more than 1-2 days of initial training for users who come with some knowledge of organic crystallography and are not afraid of the BASH command line interface. Users can always ask for our help to check the logic and progress of their workflows, but after a few weeks they hardly ever do.

Our current production version, GRACE 2.8, has been used in hundreds of crystal structure prediction studies by internal and external users. We are now preparing to release GRACE 3.1, which will represent yet another major step forward for modern crystal structure prediction.

The commercial release of GRACE currently features the following modules:

Core
The computing infrastructure.

CS​P Factory
A workflow for the fully automated execution of crystal structure prediction studies and associated tools.

Structure Solution Factory
A workflow for matching large numbers of predicted crystal structures and their mixtures to experimental powder diffraction patterns.

Force Field Factory
A workflow, plus tools to generate tailor-made force fields for third-party applications.

Well, our contract research team has been doing all of these things for many years on a daily basis.

Of course, the computational effort grows exponentially with the size, flexibility and number of independent molecules, and so the question is not so much if a project is scientifically feasible, but rather if it is economically reasonable.  Because of rapid innovation in hardware and software development, the application domain is redefined every year.

Our computational scientists will not simply dump structures and numbers on you, but will carefully work out what our computational results and your experimental results mean combined with respect to your product development and the risks you are facing.

Our contract research services are the ideal solution for new customers or those customers having to consider a few compounds per year. Even customers with a GRACE software license still use our contract research services in order to benefit from our crystallographic knowledge and our sharp analysis of solid-state problems, based on experience accumulated through hundreds of crystal structure prediction projects.