A Perspective on Unintentional Fragments and Their Impact on the Dark Metabolome, Untargeted Profiling, Molecular Networking, Public Data, and Repository Scale Analysis - PubMed (original) (raw)
Review
. 2025 Dec 1;5(12):5828-5850.
doi: 10.1021/jacsau.5c01063. eCollection 2025 Dec 22.
Affiliations
- PMID: 41450660
- PMCID: PMC12728610
- DOI: 10.1021/jacsau.5c01063
Review
A Perspective on Unintentional Fragments and Their Impact on the Dark Metabolome, Untargeted Profiling, Molecular Networking, Public Data, and Repository Scale Analysis
Yasin El Abiead et al. JACS Au. 2025.
Abstract
In/postsource fragments (ISFs) arise during electrospray ionization or ion transfer in mass spectrometry when molecular bonds break, generating ions that can complicate data interpretation. Although ISFs have been recognized for decades, their contribution to untargeted metabolomicsparticularly in the context of the so-called "dark matter" (unannotated MS or MS/MS spectra) and the "dark metabolome" (unannotated molecules)remains unsettled. This ongoing debate reflects a central tension: while some caution against overinterpreting unidentified signals lacking biological evidence, others argue that dismissing them too quickly risks overlooking genuine molecular discoveries. These discussions also raise a deeper question: what exactly should be considered part of the metabolome? As metabolomics advances toward large-scale data mining and high-throughput computational analysis, resolving these conceptual and methodological ambiguities has become essential. In this perspective, we propose a refined definition of the "dark metabolome" and present a systematic overview of ISFs and related ion forms, including adducts and multimers. We examine their impact on metabolite annotation, experimental design, statistical analysis, computational workflows, and repository-scale data mining. Finally, we provide practical recommendationsincluding a set of dos and do nots for researchers and reviewersand discuss the broader implications of ISFs for how the field explores unknown molecular space. By embracing a more nuanced understanding of ISFs, metabolomics can achieve greater rigor, reduce misinterpretation, and unlock new opportunities for discovery.
Keywords: analytical artifact; dark metabolome; electrospray ionization; in-source fragmentation; mass spectrometry; metabolomics.
© 2025 The Authors. Published by American Chemical Society.
Figures
1
One molecule, many ions. (a) Schematic view of a LC outlet and source where in-source fragments and other ions are generated. Some of the ions that are ultimately detected could also be generated within the instrument postsource ion optics. (b) Possible ion forms that might be observed (adducts, including solvent adducts, dimers, including heterodimers, and ISFs, all with isotopes). This is showcased with the bile acid, cholic acid. (c) An experimental example of a heterodimer from MSV000089018, that can occur when two molecules partially coelute and thus co-ionize. Such ion species have been shown to generally account for less than 5% of biological features in metabolomics data sets. (d) It is well-known that during ESI ATP can fragment to ADP and AMP, and ADP into AMP. Without additional experiments, these ISFs can be recognized only through chromatographic separation as ISFs will coelute with their precursor ions. Here we show that these ISFs are not always observable across three different data sets (MSV000092756 [Q Exactive Plus], MTBLS2145 [impact II UHR-TOF], and MTBLS4618 [TripleTOF 6600]). Only in one data set are ISFs clearly visible at the MS1 level consistent with the notion that ISF detection is very much experiment dependent. Moreover, it should be noted that while coelution can be used to identify ISFs here this is not as trivial when the precursors are not known like here. Direct access to plots linking to the underlying raw data are provided in the associated links.
2
Definitional flavors of the (dark) metabolome and dark matter of metabolomics. (a) The size of the dark metabolome depends on the chosen reference point, which varies with the scientific question. (b) Metabolomics data processing can be performed on the raw data file derived from a single sample, a data set containing multiple samples or a whole repository. Unique ion species can be aggregated as unique MS/MS scans or MS1 based feature extraction. What constitutes a signal is context-dependent (e.g., an MS/MS spectrum, a consensus spectrum, or an LC–MS peak). Signals can be filtered or grouped to remove bioinformatic artifacts, , uninterpretable features (such as chimeric spectra or ambiguous peaks), , and redundant ion species. However, overly strict filtering may also diminish the potential for novel discovery. The order of the workflow steps can vary depending on preference and scientific question. (c) The dark matter of metabolomics refers to the portion of metabolomics signals that can not be identified. The meaning of this concept in the context of the dark metabolome depends on analyzed samples, the metabolomics signal processing steps made, and strategies used for annotation of the observed extracted signals.
3
Sources contributing to the molecular diversity of the human metabolome. They are generally also relevant to other biological systems.
4
Strategies to group ISFs and other ion types: ISFs and adducts can be grouped in different ways as shown here for a selection of molecules. (a) It is possible to utilize peak shape correlation, which is more strict than general coelution. (b) An experimental approach to annotate ISF using in-source CID ramping: the signal intensity of the ISF will increase at 5 or 10 eV in-source CID compared to 0 eV, as demonstrated for the ISF at m/z 115.0037 from the metabolite malate (m/z 133.0142) in negative ion mode (raw data at
with accession ID MSV000087131). For instruments without in-source CID function, alternative fragmentation techniques such as all ion fragmentation (AIF) can be utilized in a similar way. (c) Enalapril, an angiotensin-converting enzyme inhibitor, detected in two samples of the data set MSV000096589 is shown after feature based molecular networking in MZmine 4. Ion species with similar MS/MS, as determined by modified cosine, are connected. The m/z value of each ion species is given on the node.
5
ISFs are molecule specific. Numbers of reported ISFs for various small molecules as reported in a standard mixture under one experimental condition by ref .
6
Leveraging beneficial aspects of ISFsISF-based (sub)structure annotation and molecular networking. (a) ISFs can suggest chemical motifs. (b) Generation of pseudo-MS/MS spectra and structure annotation. (c) An example molecular network generated using pseudo-MS/MS spectra in the IBD data set. Tyrosine-related compounds were annotated and linked. The mirror plot shows pseudo-MS/MS spectra from Tyr-C6:0 and Tyr-C8:1. Adapted from ref .
7
Multiple ions and ISF originating from the molecule often follow the same direction with respect to fold change but can vary statistically. Piperlongumine structure is shown in the upper right corner of the plot (red arrow represents where the molecule fragment in source). Horizontal dashed red lines in the plot represent a _p_-value of 0.05. Vertical dashed gray line represents down- or up-regulation by a factor of 2. The red arrow indicates where the molecules fragments to generate the ISF.
References
- Chi Y., Mitchell J. M., Zheng S., Li S.. Systematic pre-annotation explains the ‘dark matter’ in LC-MS metabolomics. bioRxiv. 2025:2025.02.04.636472. doi: 10.1101/2025.02.04.636472. -DOI
- Strachan, J. The Dark Metabolome: A Figment of Our Fragmentation? The Analytical Scientist. 2024, https://theanalyticalscientist.com/fields-applications/the-dark-metabolo....
- Strachan, J. Dark Metabolome Debate: A Call for Context. The Analytical Scientist. 2025, https://theanalyticalscientist.com/issues/2025/articles/june/dark-metabo....