Back to search

Boston University

High-throughput experiment platform development for machine learning on chemical reactivity

Abstract

dc:description.abstract

High-throughput experimentation (HTE) is a form of accelerated testing which allows for many hundreds or thousands of experiments to be conducted in parallel or in rapid sequence. Recent advances in chemical reaction miniaturization have enabled HTE application toward chemical reaction exploration, and the resultant datasets present exciting opportunities for the incorporation of machine learning (ML) with organic chemistry to expedite reaction optimization and discovery. Disclosed herein is a modular HTE chemistry reaction platform with rapid and inexpensive data acquisition capabilities for training ML algorithms on organic chemistry. Comprising almost entirely off-the-shelf components and algorithms which will be made open-source, this platform facilitates data democratization through distributed generation. Underpinning this workflow is an innovative titration-based analysis method for semi-automated and quantitative conversion data acquisition at a rate of under fifteen seconds per sample. Requisite to this platform’s success are solutions to solid and liquid reagent distribution, reaction incubation, and fast, quantitative reaction analysis which is demonstrated in a proof-of-concept screening of the SNAr reaction toward the synthesis of proteolysis targeting chimera (PROTACs). It is hoped this platform lowers the barrier for entry to HTE for chemists through its modularity, approachability, and low operating costs.

Degree

thesis:*
Name thesis:degree_name
Doctor of Philosophy
Level thesis:degree_level
doctoral
Discipline thesis:degree_discipline
Chemistry
Grantor
Boston University
Year dc:date.issued
2022

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Fraser, Douglas Gordon
Advisor dc:contributor.advisor
  • Beeler, Aaron B.
ORCID dc:identifier.orcid

Subjects

dc:subject × 7

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/2144/44797

Chain of custody

source
Harvested from
Boston University
Base URL
open.bu.edu/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Fraser, Douglas Gordon. High-throughput experiment platform development for machine learning on chemical reactivity. doctoral thesis, Boston University, 2022. https://hdl.handle.net/2144/44797