BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Using generative models to avoid rare events: insight into the the
 rmodynamics of polymorphism without sampling phase transitions - Professor
  Matteo Salvalaglio University College London
DTSTART:20251118T143000Z
DTEND:20251118T153000Z
UID:TALK240706@talks.cam.ac.uk
CONTACT:Lisa Masters
DESCRIPTION:Free energy calculations enable the quantitative understanding
  of physicochemical phenomena in material science\, chemistry\, and physic
 s. Nevertheless\, free energy methods are typically faced with computation
 al efficiency issues\, which limit their applicability in large-scale\, hi
 gh-throughput applications. One such application is the computational pred
 iction of polymorphism\, where the relative stabilities of tens to hundred
 s of putative polymorphs need to be evaluated to provide rational\, physic
 s-based prediction. A source of such limitations is that interesting metas
 table states\, representing i.e. putative polymorphs\, are usually charact
 erized by nonoverlapping configurational Boltzmann distributions\, and thu
 s\, computing free energy differences between them requires sampling inter
 mediate states characterized by high free energies and low probabilities. 
 \nIn this seminar\, I will discuss how machine learning techniques informe
 d only by locally ergodic molecular dynamics simulations can provide a blu
 eprint to boost large-scale studies of the relative thermodynamic stabilit
 y of polymorphs of molecular crystals. In particular\, we propose a combin
 ation of normalizing flow models\, and low variance free energy estimators
 1\,2 to efficiently compute the anharmonic free energy of molecular polymo
 rphs of conformationally complex organic molecules as a function of thermo
 dynamic parameters as a function of Temperature and pressure.3\,4\nReferen
 ces\n1.	Jarzynski\, C.\, (2002). Physical Review E\, 65(4)\, p.046122.\n2.
 	Olehnovics\, E.\, Liu\, Y. M.\, Mehio\, N.\, Sheikh\, A. Y.\, Shirts\, M.
  R.\, & Salvalaglio\, M. (2024). Assessing the accuracy and efficiency of 
 free energy differences obtained from reweighted flow-based probabilistic 
 generative models. Journal of Chemical Theory and Computation\, 20(14)\, 5
 913-5922.\n3.	Olehnovics\, Edgar\, et al. "Accurate Lattice Free Energies 
 of Packing Polymorphs from Probabilistic Generative Models." Journal of Ch
 emical Theory and Computation 21.5 (2025): 2244-2255.\n4.	Olehnovics\, E.\
 , Liu\, Y. M.\, Mehio\, N.\, Sheikh\, A. Y.\, Shirts\, M.\, & Salvalaglio\
 , M. (2025). Lattice free energies of molecular crystals using normalizing
  flow.
LOCATION:Todd-Hamied Room\,  Department of Chemistry
END:VEVENT
END:VCALENDAR