Disputation: Manuel Carrer

Doctoral candidate Manuel Carrer at the Department of Chemistry, Faculty of Mathematics and Natural Sciences, is defending the thesis "From All-Atom To Mesoscale: Bridging The Gap With Differentiable Molecular Dynamics" for the degree of Philosophiae Doctor.

Image may contain: A man (Manuel) is looking straight at the camera and smiling. He is wearing a black sweater, standing against a grey background.

Manuel Carrer

The Disputation will be live streamed for everyone else.
The livestream will be activated 15 minutes before the Defence starts.

Trial lecture

September 19th, 10:15 AM, Auditorium 1, Chemistry building

Trial lecture title:

"Mechanical properties of cell membranes: Helfrich theory"

 

The trial lecture will be live streamed for everyone else.
The livestream will be activated 15 minutes before the trial lecture starts.

Kreeringssammendrag/Conferral summary 

Avhandlingen studerer ulike proteinsystemer på atomistisk nivå for å hjelpe til med forståelsen av molekylære interaksjoner. Den introduserer også algoritmer og software for hybrid partikkelfelt-modellen, for å oppnå mer effektive simuleringer og parametriseringer.

Main research findings

Molecular simulations can be helpful in understanding the mechanisms at play in chemical systems, and they can provide complementary explanations to experimental results. To this aim, part of my thesis is concerned with the study of enzymes that are activated by either allosteric interactions (for example, the binding to other proteins) or post-translational modifications (such as phosphorylation). I also study the interaction of lipid bilayers with small anti-microbial peptides, that have shown the capability of disrupting bacterial membranes, and that could potentially overcome the problems of antibiotics resistance.

However, as the size of the system under study increases, the atomistic treatment incurs in substantial computational costs. Therefore, this thesis also deals with more approximate molecular representations, such as coarse-grained modeling. In particular, this work develops software to carry out hybrid particle-field simulations, where the non-bonded interaction between different particles is accounted for by a density field and not by an explicit particle-particle potential. Finally, I also propose a method for the automatic determination of hybrid particle-field parameters via a novel differentiable molecular dynamics approach, that will allow the study of biologically relevant systems in this framework.

Published Sep. 6, 2023 10:09 AM - Last modified Sep. 19, 2023 9:58 AM