Tag: UPSTAIRS

Categories
Publications

Feature article on machine learning and method development 

Feature article on machine learning and method development in liquid chromatography
Feature article on machine learning and method development in liquid chromatography

Liquid chromatography (LC) plays an important role in almost all public and private sectors. It is thus not surprising LC represents of one of the largest analytical fields in terms of resource requirements. For this reason the field is still significantly in development. This has led to the advent of high-resolution separations and multi-dimensional chromatography. However, the analytical separations technology is evolving faster than the ability of humanity to employ it effective.

Indeed, separations are often performed under suboptimal conditions and technological capabilities remain unused. Because expert knowledge and method development time are increasingly scarce, methods are often inefficient. Exploiting the full technological capabilities of liquid-phase separation technology requires deep knowledge and great time investments.

Feature article on machine learning and method development in liquid chromatography

To tackle this, Pirok started the UPSTAIRS (Unleashing the Potential of Separation Technology to Achieve Innovation in Research and Society) project in 2022 using funding from the Dutch Research Council (NWO). The goal of the project is to improve the accessibility of state-of-the-art separation technology by connecting chromatographic theory to practice using computational methods. The group aims to develop the “AutoLC” workflow in which unsupervised method development becomes possible for (2D-)LC-MS methods.

In hisf eature article on machine learning and method development, CAST researcher Gerben van Henten along with Tijmen Bos and Bob Pirok examines the role of computational methods and machine learning to facilitate automated method development. Method optimization strategies that can simultaneously optimize the large number of parameters involved are therefore of great interest to chromatographers [1].

The article focuses particularly the implementation of computer-aided workflows for the optimization of kinetic and thermodynamic parameters in LC, as well as on the possibilities to conduct this in a closed-loop fashion.

Feature article on machine learning and method development
Figure. Example of a method development workflow for unsupervised optimization described by the article [1].

In the feature article on machine learning and method development, Van Henten et al. return to the fundamental equation that quantifies resolution into three components: retention, selectivity and chromatographic efficiency. Different strategies are discussed and place into context of this distinction (Figure).

The article was published in the June edition of LC-GC Europe which was released at the largest conference for LC separations HPLC2023 in Düsseldorf, Germany. The article can be read free of charge here.

References

  1. Approaches to Accelerate Liquid Chromatography Method Development in the Laboratory Using Chemometrics and Machine Learning, G.B. van Henten, T.S. Bos, B.W.J. Pirok, LCGC Europe, 2023, 36(6), 202-209, DOI: 56530/lcgc.eu.rh7676j5.
Categories
Funding

VENI Grant for Bob Pirok

UPSTAIRS

CAST member Bob Pirok of the Van ‘t Hoff Institute for Molecular Sciences at the University of Amsterdam was awarded VENI grant in the TTW domain of the Dutch Research Council (NWO). With his grant, Pirok aims to improve the applicability of modern separation technology to enhance its relevance to society.

The Veni grants are part of the Talent Scheme of NWO and are aimed at excellent researchers who have recently obtained their doctorate. The grants of up to € 280,000 confirm the quality and innovative nature of their research and help to further establish themselves in their field over a three-year period.

The UPSTAIRS project aims to develop technology to reduce the resources required to use powerful separation systems.

The UPSTAIRS project aims to resolve this problem by developing algorithms that allow the automated development of methods using state-of-the-art instrumentation for the complex samples of this era. In UPSTAIRS, innovative peak-tracking algorithms will be developed that yield the capability of automated workflows to interpret the data. By utilizing these data for the construction of fundamental retention models for each analyte in the mixture, the algorithm can simulate numerous potential methods, regardless of the sample complexity. Novel equations will be designed for combinations of chemical method parameters. Candidate methods will be selected using novel chromatographic response functions based on Bayesian statistics and submitted for (automatic) validation by the instrument.

By Unleashing the Potential of Separation Technology for Achieving Innovation in Research and Society, UPSTAIRS will solve pressing problems and allow us to better understand materials, art, pharmaceuticals, environment, and other matrices.

Bob Pirok

Ultimately, the algorithms will enable a control computer to directly interact with the analytical instrumentation and interpret the resulting methods to then propose and evaluate a better method. According to Pirok, ‘a computer can do all this much more effectively than a human being, so that it takes far less time to develop an optimal method. This will bring the full potential of modern separation technology to society.’

Knowledge utilization is a strong focus of this project and will be achieved through a diverse and strong user committee and a demonstration of this workflow on a highly complex sample in the final work package. Moreover, a protocol will be published along with the developed open-access toolbox to allow analysts to use this work.