Monte Carlo Tree Search in Designing of High Sensitivity Derivatization Reagents For Mass Spectrometric Analysis
Name
Anneli Kruve-Viil
Abstract
Liquid chromatography electrospray mass spectrometry is widely used to detect and quantify chemicals in various samples. Unfortunately, some compounds have very low sensitivity and require derivatization before sufficient sensitivity can be achieved. Here we apply inverse molecular design, namely Monte carlo tree search, for designing derivatization reagents from a set of functional groups to improve detection of amino acids. The most efficient proved to be search with high breadth, e.g. considering 10 to 50 functional groups in each addition. We obtain reagents which are predicted to yield 50x higher sensitivity than currently used reagents. The structures of the reagents are highly novel and inspirational for the future design.
Graduation Thesis language
English
Graduation Thesis type
Master - Conversion Master in IT
Supervisor(s)
Meelis Kull
Defence year
2021