Helen L. Whelton

Non-targeted liquid chromatography tandem high-resolution mass spectrometry (LC-MS/MS) is increasingly applied for the structure-resolved chemical analysis of dissolved organic matter (DOM). With new developments in MS instrumentation and analysis software, the approach has gained substantial momentum over the past decade. However, achieving high-quality analytical data that is reproducible and comparable across laboratories can be a bottleneck in non-targeted metabolomics and organic matter chemical analysis, especially for data reuse in repository-scale analyses. Understanding the capabilities as well as challenges of comparing LC-MS/MS data from different laboratories is necessary for inferring global trends from public data sets. To illuminate instrumentation factors that drive differences and variability, we used a standardized data analysis pipeline, including classical (CMN) and feature-based molecular networking (FBMN), to analyze data from a ring trial by 24 laboratories on identical sample sets of algal and DOM extracts that were mixed in predefined concentrations and spiked with standards. Our results showed that data sets from similar mass spectrometer types with unified instrument parameters were qualitatively comparable, resolving the same general trends and shared mass spectral features. Interlaboratory comparability was best for high-intensity features, while low-intensity features showed greater detection variability. Our analysis also highlights challenges when comparing data from instruments with different acquisition rates or operating with less standardized methods. Lastly, we provide recommendations for data integration, public data sharing, standardization, and best practices for standardized LC-MS/MS data acquisition, which will be critical for long-term time series and intercomparability of DOM chemical analyses.

Non-targeted liquid-chromatography tandem high-resolution mass-spectrometry (LC-MS/MS) is increasingly applied for the structure-resolved chemical analysis of dissolved organic matter (DOM). With new developments in mass spectrometry instrumentation and analysis software, the approach has gained substantial momentum over the last decade. However, achieving high-quality analytical data that is reproducible and comparable across laboratories can be a bottleneck in non-targeted metabolomics and organic matter chemical analysis, especially for data reuse in repository-scale analyses. Understanding the capabilities as well as challenges of comparing LC-MS/MS data from different laboratories is necessary for inferring global trends from public datasets. To illuminate instrumentation factors that drive differences and variability, we used a standardized data analysis pipeline, including classical (CMN) and feature-based molecular networking (FBMN) to analyze data from a ring-trial by 24 laboratories on identical sample sets of algal and DOM extracts that were mixed in predefined concentrations and spiked with standards. Our results showed that data sets from similar mass spectrometer types with unified instrument parameters were qualitatively comparable, resolving the same general trends and shared mass spectral features. Inter-laboratory comparability was best for high intensity features, while low intensity features showed greater detection variability. Our analysis also highlights challenges when comparing data from instruments with different acquisition rates or operated with less standardized methods. Lastly, we provide recommendations for data integration, public data sharing, standardization, and best practices for standardized LC-MS/MS data acquisition, which will be critical for long-term time series and inter-comparability of DOM chemical analyses.

Cereal cultivation in Britain dates back to ca. 4000 BCE, probably introduced by migrant farmers from continental Europe. Widespread evidence for livestock appears in the archaeozoological record, also reflected by ubiquitous dairy lipids in pottery organic residues. However, despite archaeobotanical evidence for domesticated plants (such as cereals), organic residue evidence has been near-absent. Our approach, targeting low-abundance cereal-specific markers, has now revealed evidence for cereals (indicating wheat) in Neolithic pottery from Scottish ‘crannogs’, dating to ca. 3600 – 3300 BCE. Their association with dairy products suggests cereals may have been regularly prepared together as a milk-based gruel. We also observed a strong association between the occurrence of dairy products and smaller-mouthed vessels. Here, we demonstrate that cereal-specific markers can survive in cooking pots for millennia, revealing the consumption of specific cereals (wheat) that are virtually absent from the archaeobotanical record for this region and illuminating culinary traditions among early farming communities.

In European and many African, Middle Eastern and southern Asian populations, lactase persistence (LP) is the most strongly selected monogenic trait to have evolved over the past 10,000 years¹. Although the selection of LP and the consumption of prehistoric milk must be linked, considerable uncertainty remains concerning their spatiotemporal configuration and specific interactions^2,3. Here we provide detailed distributions of milk exploitation across Europe over the past 9,000 years using around 7,000 pottery fat residues from more than 550 archaeological sites. European milk use was widespread from the Neolithic period onwards but varied spatially and temporally in intensity. Notably, LP selection varying with levels of prehistoric milk exploitation is no better at explaining LP allele frequency trajectories than uniform selection since the Neolithic period. In the UK Biobank^4,5 cohort of 500,000 contemporary Europeans, LP genotype was only weakly associated with milk consumption and did not show consistent associations with improved fitness or health indicators. This suggests that other reasons for the beneficial effects of LP should be considered for its rapid frequency increase. We propose that lactase non-persistent individuals consumed milk when it became available but, under conditions of famine and/or increased pathogen exposure, this was disadvantageous, driving LP selection in prehistoric Europe. Comparison of model likelihoods indicates that population fluctuations, settlement density and wild animal exploitation—proxies for these drivers—provide better explanations of LP selection than the extent of milk exploitation. These findings offer new perspectives on prehistoric milk exploitation and LP evolution.

Chemical analyses of carbonized and absorbed organic residues from archaeological ceramic cooking vessels can provide a unique window into the culinary cultures of ancient people, resource use, and environmental effects by identifying ingredients used in ancient meals. However, it remains uncertain whether recovered organic residues represent only the final foodstuffs prepared or are the accumulation of various cooking events within the same vessel. To assess this, we cooked seven mixtures of C₃and C₄foodstuffs in unglazed pots once per week for one year, then changed recipes between pots for the final cooking events. We conducted bulk stable-isotope analysis and lipid residue analysis on the charred food macro-remains, carbonized thin layer organic patina residues and absorbed lipids over the course of the experiment. Our results indicate that: (1) the composition of charred macro-remains represent the final foodstuffs cooked within vessels, (2) thin-layer patina residues represent a mixture of previous cooking events with bias towards the final product(s) cooked in the pot, and (3) absorbed lipid residues are developed over a number of cooking events and are replaced slowly over time, with little evidence of the final recipe ingredients.

We unequivocally confirmed pre-Clovis human occupation of the western Great Basin using fecal lipid biomarkers and a new 14C date.