MealTime-MS: A Machine Learning-Guided Real-Time Mass Spectrometry Analysis for Protein Identification and Efficient Dynamic Exclusion
Bibliographic record
Abstract
ABSTRACT Mass spectrometry-based proteomics technologies are the prime methods for the high-throughput identification of proteins in complex biological samples. Nevertheless, there are still technical limitations that hinder the ability of mass spectrometry to identify low abundance proteins in complex samples. Characterizing such proteins is essential to provide a comprehensive understanding of the biological processes taking place in cells and tissues. Still today, most mass spectrometry-based proteomics approaches use a data-dependent acquisition strategy, which favors the collection of mass spectra from proteins of higher abundance. Since the computational identification of proteins from proteomics data is typically performed after mass spectrometry analysis, large numbers of mass spectra are typically redundantly acquired from the same abundant proteins, and little to no mass spectra are acquired for proteins of lower abundance. We therefore propose a novel supervised learning algorithm that identifies proteins in real-time as mass spectrometry data are acquired and prevents further data collection from confidently identified proteins to ultimately free mass spectrometry resources to improve the identification sensitivity of low abundance proteins. We use real-time simulations of a previously performed mass spectrometry analysis of a HEK293 cell lysate to show that our approach can identify 92.1% of the proteins detected in the experiment using 66.2% of the MS2 spectra. We also demonstrate that our approach outperforms a previously proposed method, is sufficiently fast for real-time mass spectrometry analysis, and is flexible. Finally, MealTime-MS’ efficient usage of mass spectrometry resources will provide a more comprehensive characterization of proteomes in complex samples.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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".