Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

In Analytical chemistry

About ten years ago most of the liquid chromatography (LC) electrospray (ESI) mass spectrometry (MS) analysis of environmental, metabolomics, pharmaceutical samples were predominantly carried out as targeted analysis. Targeted analyses allow detection and quantification a few selected analytes with the aid of standard substances. Today, the centre of gravity is shifting towards non-targeted methods which utilize high-resolution mass spectrometry (HRMS). Furthermore, the targeted and non-targeted methods are merging into each other. LC/HRMS based non-targeted methods allow detecting compounds recovered from the sample preparation and ionizing in the electrospray ionization (ESI) source. New possibilities arising from applying the machine learning tools to LC/HRMS data have already transformed the process of identifying the compounds. The computer-aided identification process is not compatible with traditional calibration graph based quantification methods. The main obstacle arises from the fact that in ESI different compounds ionize to a very different extent. The differences up to 100 million times have been reported. This phenomenon results in a vastly different response of different compounds at the same concentration and complicates the quantification for compounds without standard substances. However, decision making is hindered without quantitative information. Therefore, the need to obtain quantitative information from the non-targeted analysis is triggering an emerging field of research. This review aims at giving an overview of different possibilities for quantitatively comparing the results obtained from LC/HRMS based non-targeted analysis. More specifically, quantification via structurally similar internal standards, different isotope labelling strategies, radiolabelling, and predicted ionization efficiencies are reviewed.

Kruve Anneli