FlexCryst Case Studies

Molecular Modeling with FlexCryst (Overview)

This case study explores the use of data from the Cambridge Structure Database (CSD) to predict the physical and chemical properties of unknown substances. It discusses how FlexCryst, with its powerful computational facilities, can be used to derive Gibbs free energy and predict properties like phase transitions and solubility, highlighting its impact on pharmaceutical research.

Crystal Structure Engineering with FlexCryst by Visualization of the Intermolecular Interactions

This study introduces a data mining approach to crystallography that assigns weights to different descriptors, moving beyond traditional statistical estimations. It presents a force field trained on 100,000 structures, enabling the visualization of intermolecular forces to identify structural inaccuracies and inform targeted modifications to achieve desired properties, such as a denser packing for explosives or higher solubility for drugs.

Prediction of Metal Coordination by Data Mining of experimental Crystal Structures

This case study tackles the complex problem of predicting organometallic crystal structures by including the prediction of metal coordination. It details a multi-step data mining procedure on the CSD to parameterize effective potentials, demonstrating how the method can predict the correct coordination for complexes, using nickel-complexes as a practical example.

From Crystals by Data Mining to Molecular Modeling

Leveraging the vast data from crystallographic databases, this study explores the application of supervised and unsupervised classification. It demonstrates how these tools can be used to identify polymorphs and derive accurate interaction potentials that go beyond quantum chemistry methods, enabling the prediction of effects like isotopic effects and the temperature dependence of crystal structures.

A new similarity index for the crystal structure determination and cluster analysis of powder diagrams

This case study presents a novel similarity index for the automated comparison of powder diagrams. Unlike traditional methods, this new index remains effective even with large deviations in cell constants, facilitating the determination of crystal structures from un-indexed powder diagrams, as demonstrated with organic pigments.

Solubility and Crystal Structure Prediction

Using Data Mining Force Fields (DMFF), this study shows a fast and accurate method for predicting the solubility of drugs. By calculating the free energy of the crystal lattice and correlating it with hydration energy, the model achieves a high correlation with experimental solubility data, making it a powerful tool for drug development.

Lattice energy calculation: a quick tool for screening of stability and relative solubility of co-crystals

Focusing on co-crystals, this case study demonstrates how FlexCryst's data mining force field can be used for in silico screening. By estimating the relative stability and solubility of co-crystals, the study proves the utility of the program in tailoring the properties of active pharmaceutical ingredients (APIs).

Data Mining Force Field: Crystal Properties Estimation and Screening for Polymorphs, Co-crystals and Salts

This study presents an improved data mining approach for force fields, which allows for the screening of large datasets to find polymorphs, co-crystals, and salts. It highlights the force field's accuracy in estimating energy and its perfect correlation with experimental data, even for challenging classes of compounds like carboxylic and amino acids, and its success in predicting a high percentage of crystal structures within a top energy rank.