Baseline-referenced plasma metabolomic trajectories during early post-exercise recovery: A secondary analysis of public human exercise metabolomics data.
A baseline-referenced secondary analysis of public plasma metabolomics datasets organized early post-exercise responses by baseline return, residual deviation, external support, and intensity-related behavior, and prioritized candidate metabolites that should be treated as hypothesis-generating rather than validated recovery biomarkers.
Key Findings
Results
In the primary discovery dataset ST001789, a large number of plasma metabolites differed from baseline immediately after a 90-min endurance run and many remained different after 60 min of recovery.
198 metabolites differed from baseline immediately after exercise.
140 metabolites differed from baseline after 60 min of recovery.
116 metabolites changed between the immediate post-exercise and recovery samples.
The dataset included 57 paired plasma samples from 19 participants (pre-exercise, immediately post-exercise, and 60 min recovery).
Results
Recovery-pattern classification showed that early post-exercise metabolic responses were heterogeneous.
Metabolites were classified into recovery-pattern classes using baseline-referenced contrasts.
The analysis distinguished transient exercise responses from residual baseline deviation at 60 min of recovery.
Results
Most discovery features that could be matched in the external dataset ST004303 showed direction-consistent and FDR-supported changes.
Of 77 discovery features matched in ST004303, 74 had at least one direction-consistent external comparison.
66 of the 77 had at least one FDR-supported external change.
ST004303 included moderate-intensity and sprint-interval exercise conditions, allowing assessment of intensity-related behavior.
Results
Staged prioritization identified 54 primary candidate metabolites for future validation.
Candidates were intended to prioritize metabolites for future validation.
Results
Threshold sensitivity analyses showed that the candidate set shrinks under stricter corrected schemes, with a small core retained across all schemes.
Corrected sensitivity schemes retained 136, 25, and 11 features.
11 features were retained across all three schemes.
Results
The prioritized candidate metabolites spanned several biological themes.
Themes included fatty-acid transport and oxidation, amino acid metabolism, lipid remodeling or signaling, and purine/nucleoside turnover.
Conclusions
Baseline return, residual deviation, external support, and intensity-related behavior help organize early post-exercise metabolomic responses, but the identified metabolites are hypothesis-generating rather than validated biomarkers.
The study was a secondary analysis of two public human plasma metabolomics datasets.
The primary dataset was small (19 participants), and external support came from a separate dataset with different exercise conditions.
The authors state the metabolites are not validated recovery biomarkers.
What This Means
This study re-analyzed two publicly available datasets of blood metabolites measured around exercise. The main dataset had 19 people who gave blood before a 90-minute run, right after it, and after an hour of recovery. Instead of just looking at what changed after exercise, the authors compared each time point to the person's own pre-exercise baseline to separate short-lived exercise effects from changes that were still present after an hour of rest.
Many metabolites changed right after the run (198), and a large number (140) were still different from baseline after 60 minutes, showing that recovery in the first hour is incomplete and varies between metabolites. Of the 77 metabolites that could be matched in a second dataset with moderate and sprint-interval exercise, most showed changes in a consistent direction. The authors narrowed the list to 54 candidates, and only 11 held up under all of their stricter statistical checks. These involved fat transport and burning, amino acids, lipid signaling, and breakdown of energy-related compounds.
This research suggests that comparing to baseline is a useful way to interpret early post-exercise metabolic changes. However, the sample was small and the work reused existing data, so the identified metabolites are leads for future studies, not proven markers of recovery.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Feng Y, Wu Z. (2026). Baseline-referenced plasma metabolomic trajectories during early post-exercise recovery: A secondary analysis of public human exercise metabolomics data.. PloS one. https://doi.org/10.1371/journal.pone.0361757