Multilaboratory Untargeted Mass Spectrometry Metabolomics Collaboration to Identify Bottlenecks and Comprehensively Annotate A Single Dataset
Bibliographic record
Abstract
High Resolution Image Download MS PowerPoint Slide Annotation is the process of assigning features in mass spectrometry metabolomics data sets to putative chemical structures or “analytes.” The purpose of this study was to identify challenges in the annotation of untargeted mass spectrometry metabolomics datasets and suggest strategies to overcome them. Toward this goal, we analyzed an extract of the plant ashwagandha ( Withania somnifera ) using liquid chromatography–mass spectrometry on two different platforms (an Orbitrap and Q-ToF) with various acquisition modes. The resulting 12 datasets were shared with ten teams that had established expertise in metabolomics data interpretation. Each team annotated at least one positive ion dataset using their own approaches. Eight teams selected the positive ion mode data-dependent acquisition (DDA) data collected on the Orbitrap platform, so the results reported for that dataset were chosen for an in-depth comparison. We compiled and cross-checked the annotations of this dataset from each laboratory to arrive at a “consensus annotation,” which included 142 putative analytes, of which 13 were confirmed by comparison with standards. Each team only reported a subset (24 to 57%) of the analytes in the consensus list. Correct assignment of ion species (clusters and fragments) in MS spectra was a major bottleneck. In many cases, in-source redundant features were mistakenly considered to be independent analytes, causing annotation errors and resulting in overestimation of sample complexity. Our results suggest that better tools/approaches are needed to effectively assign feature identity, group related mass features, and query published spectral and taxonomic data when assigning putative analyte structures.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.053 | 0.048 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.002 |
| Bibliometrics | 0.006 | 0.005 |
| Science and technology studies | 0.004 | 0.002 |
| Scholarly communication | 0.007 | 0.007 |
| Open science | 0.005 | 0.020 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.003 | 0.005 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".