Integrating small-angle neutron scattering with machine learning enhances the measurement of complex molecular structures.

Small angle scattering (SAS) is a powerful technique for studying nanoscale samples, but its use in research has been hindered until now because it requires prior knowledge of the sample's chemical composition. European Journal of PhysicsEugen Anitas from the Bogolyubov Institute of Theoretical Physics in Dubna, Russia, will present a more advanced approach to integrating SAS with machine learning algorithms.

The technique, named α-SAS, can analyze molecular samples without the need for extensive preparation or computational resources, enabling researchers to gain greater insight into the properties of complex biomolecules such as proteins, lipids and carbohydrates.

SAS measures the deflection of radiation (usually X-rays or neutrons) after it interacts with molecular structures suspended in a solvent. By adjusting the composition of the solvent, researchers can increase or decrease the visibility of certain components of the system, a technique called “contrast variation.” However, for this technique to work, researchers need knowledge of the chemical composition of the sample before starting the experiment.

Through his research, Anitas overcame this limitation by integrating SAS with machine learning algorithms, creating a method called α-SAS, which estimates small-angle neutron scattering (SANS) results by running many random simulations of suspended samples and analyzing the distribution of the results.

Anitas demonstrated the capabilities of α-SAS through two different case studies. In the first, he investigated “Janus particles,” artificial self-propelled structures with well-known changes in contrast and neutron scattering intensity. In the second, he tested the technique on a complex protein-based molecular system.

In both cases, Anitas was able to determine the molecular structures much more efficiently than he would have done without the integration of machine learning. Based on these promising results, Anitas now hopes that his approach will make SAS an even more powerful tool for analyzing molecular structures.