doi: 10.1038/s41526-023-00337-5.
Affiliations
Free PMC article
Item in Clipboard
NPJ Microgravity.
.
Free PMC article
Abstract
The adverse effects of microgravity exposure on mammalian physiology during spaceflight necessitate a deep understanding of the underlying mechanisms to develop effective countermeasures. One such concern is muscle atrophy, which is partly attributed to the dysregulation of calcium levels due to abnormalities in SERCA pump functioning. To identify potential biomarkers for this condition, multi-omics data and physiological data available on the NASA Open Science Data Repository (osdr.nasa.gov) were used, and machine learning methods were employed. Specifically, we used multi-omics (transcriptomic, proteomic, and DNA methylation) data and calcium reuptake data collected from C57BL/6 J mouse soleus and tibialis anterior tissues during several 30+ day-long missions on the international space station. The QLattice symbolic regression algorithm was introduced to generate highly explainable models that predict either experimental conditions or calcium reuptake levels based on multi-omics features. The list of candidate models established by QLattice was used to identify key features contributing to the predictive capability of these models, with Acyp1 and Rps7 proteins found to be the most predictive biomarkers related to the resilience of the tibialis anterior muscle in space. These findings could serve as targets for future interventions aiming to reduce the extent of muscle atrophy during space travel.
© 2023. This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply.
Conflict of interest statement
The authors declare no competing interests.
Figures
Fig. 1. QLattice regression analysis of TA multi-omics data and calcium uptake.
a Representative examples of the mathematical relationships between multi-omic features identified by QLattice to predict calcium reuptake in TA muscle during LOOCV. b Top 9 features ranked by how many times they were used in a model found by QLattice during LOOCV. c T1 and T10 cross-validated R2 scores, as well as the number of RNA-seq and proteomic features that were found among the top 50 features. d Gene set enrichment analysis results using the top nine genes from the QLattice analysis.
Fig. 2. QLattice regression analysis of SOL multi-omics data and calcium uptake.
a Representative examples of the mathematical relationships between multi-omic features identified by QLattice to predict calcium reuptake in SOL muscle during LOOCV. b Top 13 features ranked by how many times they were used in a model found by QLattice during LOOCV. c T1 and T10 cross-validated R2 scores, as well as the number of RNA-seq and methylation features that were found among the top 50 features. d Gene set enrichment analysis results using the top 13 genes from the QLattice analysis.
Fig. 3. Gene/protein relationship with calcium reuptake in SOL and TA muscles and putative mechanism.
The expression levels of the top key genes identified by QLattice analysis and their corresponding protein levels are shown against the calcium reuptake AUC in both TA and SOL muscles flown in space (FLT) or from ground controls (GC). a QLattice key protein levels in TA. b QLattice key gene expression levels in TA. c QLattice key gene expression levels in SOL. d Putative mechanism based on the Acyp1 response, showing up in the majority of the models for TA muscle and e SOL muscle. f Calcium reuptake AUC in TA muscle. g Calcium reuptake AUC in SOL muscle. All significance was calculated using Mann–Whitney–Wilcoxon test two-sided: *: 1.00e−02 < p ≤ 5.00e−02, **:1.00e−03 < p ≤ 1.00e−02).
Fig. 4. QLattice classification analysis of TA multi-omics data and FLT/GC groups.
a Representative examples of the mathematical relationships between multi-omic features identified by QLattice to predict FLT versus GC in TA muscle during LOOCV. b Top 11 features ranked by how many times they were used in a model found by QLattice during LOOCV. c T1 and T10 cross-validated R2 scores, as well as the total number of RNA-seq, proteomics, and methylation features across all models. d Gene set enrichment analysis results using the top 11 features from the QLattice analysis.
Fig. 5. Expression levels of top proteins and genes identified by QLattice TA and SOL classification analysis and muscle weights.
a QLattice key protein levels in TA. b QLattice key gene levels in TA. c QLattice key gene levels in SOL. All significance was calculated using Mann–Whitney–Wilcoxon test two-sided: *: 1.00e−02 < p ≤ 5.00e−02, **:1.00e−03 < p ≤ 1.00e−02).
Fig. 6. QLattice classification analysis of SOL multi-omics data and FLT/GC groups.
a Representative examples of the mathematical relationships between multi-omic features identified by QLattice to predict FLT versus GC in SOL muscle during LOOCV. b Top 9 features ranked by how many times they were used in a model found by QLattice during LOOCV. c T1 and T10 cross-validated R2 scores, as well as the total number of RNA-seq and methylation features across all models. d Gene set enrichment analysis results using the top 9 features from the QLattice analysis.
References
LinkOut – more resources
-
Full Text Sources
-
Miscellaneous
