Seminar by Arthi Jayaraman, Ph.D University of Delaware, Newark DE

Machine Learning and Computational (CREASE) Analysis of Small Angle X-ray Scattering Data from Supramolecular Dipeptide Systems

in ILL50, Room 110/111

---

My research group’s expertise lies in the development of physics-based molecular models and simulation methods as well as data-driven machine learning models for designing and characterizing soft macromolecular materials. We have devoted significant efforts towards the development of machine learning based computational methods to accelerate and automate interpretation of structural characterization data from scattering and microscopy techniques (e.g., CREASE and CREASE-2D [1-3], PairVAE [4], microscopy analyses [5,6]). All open-source codes can be accessed here: https://github.com/arthijayaraman-lab

In this talk I will present our most recent work using machine learning based CREASE-2D method to analyze two-dimensional (2D) scattering patterns obtained from small angle X-ray scattering (SAXS) measurements of supramolecular peptide micellar systems. Traditional analysis methods of 2D SAXS patterns involve fitting approximate or incorrect analytical models to azimuthally-averaged 1D scattering patterns where azimuthal averaging can miss key anisotropic structural arrangements. Analysis of the 2D scattering profiles of such micellar solutions using CREASE-2D allows us to understand both isotropic and anisotropic structural arrangements that are present in systems of assembled dipeptides in water and in the presence of added solvents/salts. CREASE-2D outputs distributions of relevant structural features including ones that cannot be identified with existing analytical models (e.g., assembled tubes’ cross-sectional eccentricity, tubes’ orientational order). CREASE-2D also provides representative three-dimensional (3D) real-space structural representations for the optimized values of these structural features to facilitate visualization of the structures. Through this detailed interpretation of these 2D SAXS profiles we are able to characterize the orientational order and shapes of the assembled tube structures as a function of dipeptide chemistry, solution conditions with varying salts and solvents, and relative concentrations of all components.

References

[1] C. M. Heil et al.,  ACS Central Science 8, 7, 996-1007 (2022).

[2] C. M. Heil et al., JACS Au 3, 3, 889–904 (2023).

[3] S.V.R. Akepati et al., JACS Au  4, 4, 1570–1582 (2024).

[4] S. Lu and A. Jayaraman, JACS Au 3, 9, 2510–2521 (2023).

[5] A. Paruchuri et al., Digital Discovery, 3, 2533-2550 (2024)

[6] S. Lu and A. Jayaraman, Progress in Polymer Science  153, 101828 (2024)

Biography

Arthi Jayaraman is currently a full professor in the Departments of Chemical and Biomolecular Engineering and Materials Science and Engineering at the University of Delaware (UD), Newark. She is also the director for an NSF-funded NRT graduate traineeship program on ​‘Computing and Data Science Training for Materials Innovation, Discovery, and Analytics’

She received her Ph.D. in Chemical Engineering from North Carolina State University and conducted her postdoctoral research in Materials Science and Engineering at the University of Illinois-Urbana Champaign.

Jayaraman’s research has been recognized with honors/awards including AIChE COMSEF IMPACT Award (2021), Fellow of the APS (2020), ACS PMSE Young Investigator (2014), and AIChE COMSEF Young Investigator Award (2013), and Department of Energy (DOE) Early Career Research Award (2010

Honors recognizing Jayaraman’s excellence in teaching include Univ. of Delaware College of Engineering faculty award for excellence in teaching (2023), Univ. of Colorado chemical and biological engineering’s outstanding graduate teaching award (2014), and Univ. of Colorado chemical and biological engineering’s outstanding undergraduate teaching award (2011).