Biology subjects

Costes, S. V.

Publications and source records attributed to Costes, S. V..

10 recordsLinked to original sources

bioRxiv2026

A systematic imputation framework for sparse, multimodal space biology datasets: application to retinal imaging and omics from the RR9 mission

Missing data is a fundamental challenge in space biology, where high experimental costs, limited sample availability, and tissue allocation constraints produce datasets that are sparse, multimodal, and heterogeneous. We present a systematic four-stage framework for diagnosing, implementing, and validating data imputation strategies tailored to these characteristics, and demonstrate its application to retinal imaging and omics data from the NASA Rodent Research 9 (RR9) mission. Using logistic regression-based missingness diagnosis, we identify a Missing At Random (MAR) mechanism driven by experimental design constraints across nine assay modalities. We implement and optimize three imputation strategies: K-Nearest Neighbors (KNN), Multiple Imputation by Chained Equations with weak ElasticNet regularization (MICE-Elastic), and a per-column hybrid strategy, evaluated against a random sample imputer baseline. Validation across seven complementary metrics including supervised classification, unsupervised clustering, correlation structure preservation, masked value recovery, cross-dataset generalization, and permutation testing reveals that MICE-Elastic and the Hybrid strategy preserve genuine biological signal in both RNA-seq and TUNEL modalities, while KNN and the random sample imputer do not despite achieving comparable cross-validation accuracy. A critical finding is that imputation substantially improves supervised classification performance while consistently degrading unsupervised clustering structure, a trade-off researchers must understand before applying these methods. This framework provides practical, actionable guidance for space biologists and data scientists managing sparse multimodal datasets, and represents a foundational step toward digital twin development for space medicine.

bioinformatics↗

bioRxiv2026

Cross-Species Adaptation of RETFound for Rodent OCT Age Estimation Reveals Strong CNN Baselines in Data-Scarce Space Biology

Space-biology imaging studies are often constrained by severe data scarcity, limiting the development of robust machine-learning biomarkers. Rodent spaceflight and space-analog datasets provide an important preclinical setting for testing transfer-learning strategies, but the extent to which human retinal foundation models can generalize to rodent optical coherence tomography (OCT) remains unclear. Here, we benchmark cross-species adaptation of RETFound, a human retinal Vision Transformer pretrained on 1.6 million retinal images, for chronological age prediction from Brown Norway rat OCT B-scans in the NASA Open Science Data Repository dataset OSD-679. We adapted RETFound using Low-Rank Adaptation (LoRA) and evaluated performance on control animals under matched 3-fold rat-level cross-validation. We compared RETFound+LoRA with a strong ImageNet-pretrained Xception baseline under matched protocols and included a scratch/random ViT as a supplementary negative-control architecture check. Metrics included mean absolute error (MAE), R2, and inter-eye mean absolute difference (MAD). RETFound+LoRA achieved MAE = 26.20 {+/-} 5.03 days with R2 = 0.744 {+/-} 0.049. However, Xception performed better in the primary benchmark (MAE = 19.01 {+/-} 7.67 days, R2 = 0.853 {+/-} 0.082), and the matched-fold comparison favored Xception, although this result should be interpreted cautiously given the small number of folds. Inter-eye consistency was maintained across the matched control evaluation, and saliency maps localized model attention to anatomically plausible inner retinal regions. Together, these results show that human retinal foundation models can transfer to rodent OCT in a scientifically useful way, but also that strong CNN baselines may outperform transformer-based models in small-sample cross-species settings. This preprint provides a reproducible benchmark and baseline framework for future retinal biomarker development in space biology. Significance StatementSpace-biology imaging studies are intrinsically data-limited. This preprint provides a reproducible cross-species benchmark for adapting Earth-trained retinal models to rodent OCT under small-sample conditions, highlights the value of strong CNN baselines, and offers a reusable starting point for future retinal biomarker development in space-relevant datasets.

bioinformatics↗

bioRxiv2025

Machine learning ensemble identifies distinct age-related response to spaceflight in mammary tissue

BackgroundSpaceflight presents unique environmental stressors, such as microgravity and radiation, that significantly affect biological systems at the molecular, cellular, and organismal levels. Astronauts face an increased risk of developing cancer due to exposure to ionizing radiation and other spaceflight-related factors. Age plays a crucial role in the bodys response to the cellular stresses that lead to cancer, with younger organisms generally exhibiting more efficient response mechanisms than older ones. The vast majority of research investigating breast cancer risk from spaceflight is done in vitro, using cell lines exposed to simulated radiation and microgravity. ObjectivesThe primary objective of this observational study is to characterize the response to spaceflight of in vivo murine mammary tissue and identify the molecular biomarkers enriched in this response using mice flown on the International Space Station. The secondary objective is to determine if age plays a role in this response. MethodsThe NASA Open Science Data Repository (OSDR) has curated transcriptomic data obtained from murine mammary tissue in a controlled experiment (OSD-511) which includes 43 young and old female mice. In this study, we utilized an ensemble of four machine learning binary classifiers (logistic regression, support vector machine, random forest, and single-layer perceptron) to analyze gene expression profiles to predict age (old vs young) and condition (spaceflight vs ground control). Using the genes our ensemble identified as most predictive, we performed pathway enrichment analysis to investigate the molecular pathways involved in spaceflight-related health risks, particularly in the context of breast cancer. ResultsAll space-flown mice responded to spaceflight with evidence of systemic metabolic reprogramming and mitochondrial adaptation to microgravity and radiation as compared to their 33 ground control counterparts. For the 10 mice flown in space, older mice exhibited chronic, significantly enriched pathways related to cell adhesion and extracellular matrix (ECM) structure, while younger mice showed acute activation of pathways involved in cortisol synthesis and adrenergic stress response (with false discovery rate adjusted q-values < .001). These results provide putative biomarkers for monitoring and treating breast cancer. ConclusionsThese findings highlight the critical role of age in modulating the response to spaceflight-induced stress and suggest that these molecular pathways may contribute to differential outcomes in tissue homeostasis, metabolic disorders, and breast cancer tumorigenesis. Moreover, our computational methodology may be applied to several unexplored datasets in OSDR and beyond.

bioinformatics↗

bioRxiv2025

Foundational Architecture Enabling Federated Learning for Training Space Biomedical Machine Learning Models between the International Space Station and Earth

The public and commercial space industries are planning longer duration and more distant space missions, including the establishment of a habitable lunar base and crewed missions to Mars. To support Earth-independent scientific and medical operations, such missions can leverage artificial intelligence and machine learning models to assist with crew healthcare, spacecraft maintenance, and other critical tasks. However, transferring large volumes of data between Earth and space for model development consumes valuable bandwidth, is vulnerable to communication disruptions, and may compromise crew safety and data privacy. Federated learning enables model training while keeping data in situ and only transferring model parameters. In this work, we present a flexible, resilient federated learning framework that provides the secure transmission of model updates between Earth and the International Space Station. On March 15, 2024, this framework pioneered the deployment of federated learning in a spaceflight setting, training classifier models between Earth and the ISS using both real biomedical research data and synthetically generated data.

bioinformatics↗

bioRxiv2024

A Machine Learning Model of Perturb-Seq Data for Use in Space Flight Gene Expression Profile Analysis

The genetic perturbations caused by spaceflight on biological systems tend to have a system-wide effect which is often difficult to deconvolute into individual signals with specific points of origin. Single cell multi-omic data can provide a profile of the perturbational effects but does not necessarily indicate the initial point of interference within a network. The objective of this project is to take advantage of large scale and genome-wide perturbational or Perturb-Seq datasets by using them to pre-train a generalist machine learning model that is capable of predicting the effects of unseen perturbations in new data. Perturb-Seq datasets are large libraries of single cell RNA sequencing data collected from CRISPR knock out screens in cell culture. The advent of generative machine learning algorithms, particularly transformers, make it an ideal time to re-assess large scale data libraries in order to grasp cell and even organism-wide genomic expression motifs. By tailoring an algorithm to learn the downstream effects of the genetic perturbations, we present a pre-trained generalist model capable of predicting the effects of multiple perturbations in combination, locating points of origin for perturbation in new datasets, predicting the effects of known perturbations in new datasets, and annotation of large-scale network motifs. We demonstrate the utility of this model by identifying key perturbational signatures in RNA sequencing data from spaceflown biological samples from the NASA Open Science Data Repository.

bioinformatics↗

bioRxiv2024

Systemic Genome Correlation Loss as a Central Characteristic of Spaceflight

Spaceflight exposes the human body to a unique combination of stressors--microgravity, radiation, and confinement--that induces multisystemic physiological dysregulation. Traditional transcriptomic analyses have focused on differential expression to identify key genes, yet this approach fails to explain why astronauts experience systemic fragility despite often subtle changes in gene abundance. Here, we present a comprehensive meta-analysis of 10 independent transcriptomic and genomic datasets (N = 136) from the NASA Open Science Data Repository. By shifting focus from gene abundance to gene-gene correlation topology, we identify Systemic Genome Correlation Loss as a central biosignature of spaceflight. We show that the regulatory architecture of the transcriptome undergoes a profound decoherence in microgravity, shifting the global correlation distribution toward stochasticity (p < 10-15). This phenomenon is universal across tissues and independent of gene variance. We identify a massive population of 760 genes that maintain stable expression levels but lose over 500 regulatory connections each, outnumbering canonical differentially expressed genes by three-to-one. Finally, we demonstrate that cells preserve the connectivity of survival-critical DNA repair networks preferentially while allowing mitochondrial and synaptic networks to shatter. These findings suggest that astronaut health risks are driven by the entropic decay of regulatory synchronization, proposing a new paradigm for countermeasure development focused on network stabilization rather than pathway inhibition.

genomics↗

bioRxiv2022

Mouse Genomic Associations with Ex Vivo Sensitivity to Simulated Space Radiation

Exposure to ionizing radiation is considered by NASA to be a major health hazard for deep space exploration missions. Ionizing radiation sensitivity is modulated by both genomic and environmental factors. Understanding their contributions is crucial for designing experiments in model organisms, evaluating the risk of deep space (i.e. high-linear energy transfer, or LET, particle) radiation exposure in astronauts, and also selecting therapeutic irradiation regimes for cancer patients. We identified single nucleotide polymorphisms in 15 strains of mice, including 10 collaborative cross model strains and 5 founder strains, associated with spontaneous and ionizing radiation-induced ex vivo DNA damage quantified based on immunofluorescent 53BP1+ nuclear foci. Statistical analysis suggested an association with pathways primarily related to cellular signaling, metabolism, tumorigenesis and nervous system damage. We observed different genomic associations in early (4 and 8 hour) responses to different LET radiation, while later (24 hour) DNA damage responses showed a stronger overlap across all LETs. Furthermore, a subset of pathways was associated with spontaneous DNA damage, suggesting 53BP1+ foci as a potential biomarker for DNA integrity in mouse models. Based on our results, we suggest several mouse strains as new models to further study the impact of ionizing radiation and validate the identified genetic loci. We also highlight the importance of future human ex vivo studies to refine the association of genes and pathways with the DNA damage response to ionizing radiation and identify targets for space travel countermeasures.

genomics↗

bioRxiv2021

The Great Deceiver: miR-2392's Hidden Role in Driving SARS-CoV-2 Infection

MicroRNAs (miRNAs) are small non-coding RNAs involved in post-transcriptional gene regulation that have a major impact on many diseases and provides an exciting avenue towards antiviral therapeutics. From patient transcriptomic data, we have discovered a circulating miRNA, miR-2392, that is directly involved with SARS-CoV-2 machinery during host infection. Specifically, we show that miR-2392 is key in driving downstream suppression of mitochondrial gene expression, increasing inflammation, glycolysis, and hypoxia as well as promoting many symptoms associated with COVID-19 infection. We demonstrate miR-2392 is present in the blood and urine of COVID-19 positive patients, but not detected in COVID-19 negative patients. These findings indicate the potential for developing a novel, minimally invasive, COVID-19 detection method. Lastly, using in vitro human and in vivo hamster models, we have developed a novel miRNA-based antiviral therapeutic that targets miR-2392, significantly reduces SARS-CoV-2 viability in hamsters and may potentially inhibit a COVID-19 disease state in humans.

systems biology↗

bioRxiv2020

DNA break clustering inside repair domains predicts cell death and mutation frequency in human fibroblasts and in Chinese hamster cells for a 10^3x range of linear energy transfers

Cosmic radiation, composed of high charged and energy (HZE) particles, causes cell death and mutations that can subsequently lead to cancers. Radiation-mediated mutations are induced by inter- and intra-chromosomal rearrangements (translocations, deletions, inversions) that are triggered by misrepaired DNA breaks, especially double-strand breaks (DSBs). In this work, we introduce a new model to predict radiation-mediated induction of cell death and mutation in two different cell lines across a large range of linear energy transfer (LET) values, based on the assumption that DSBs cluster into repair domains, as previously suggested by our group. Specifically, we propose that the probabilities of cell survival and cell mutation can be determined from the number of DSBs and the number of pairwise DSB interactions forming radiation-induced foci. We computed the distribution and locations of DSBs with the new simulation code RITCARD (relativistic ion tracks, chromosome aberrations, repair, and damage) and combined them with experimental data from HF19 human fibroblasts and V79 Chinese hamster cells to derive the parameters of our model and expand its predictions to the relative biological effectiveness (RBE) for cell survival and mutation in both cell lines in response to 9 different irradiation particles and energies ranging from 10 to 1,600 MeV/n. Our model generates the correct bell shape of LET dependence for RBE, as well as similar RBE values as experimental data, notably including data that were not used to set the model parameters. Interestingly, our results also suggest that cell orientation (parallel or perpendicular) with respect to the HZE beam can modulate the RBE for both cell death and mutation frequency. Cell orientation effects, if confirmed experimentally, would be another strong piece of evidence for the existence of DNA repair domains and their critical role in interpreting cellular sensitivity to cosmic radiation and hadron therapy. AUTHOR SUMMARYOne of the main hazards of human spaceflight beyond low Earth orbit is space radiation exposure. Galactic cosmic rays (GCRs), in particular their high-charge and high-energy particle component, induce a unique spatial distribution of DNA double strand breaks in the nucleus along their traversal in the cell [1], which result in significantly higher cancer risk than X-rays [2]. To mitigate this hazard, there is a significant need to better understand and predict the effects of cosmic radiation exposure at the cellular level. We have computationally predicted two biological endpoints - cell survival and probability of mutations, critical for cancer induction mechanisms - for the full spectrum of cosmic radiation types and energies, by modeling the distribution of DNA damage locations within the cell nucleus. From experimental results of cell survival and mutation probability in two standard cell lines, we were able to derive the parameters of the model for multiple radiation qualities, both biological endpoints, and two irradiation orientations. The model was validated against biological data and showed high predictive capability on data not used for tuning the model. Overall, this work opens new perspectives to predict multiple responses to cosmic radiation, even with limited experimental data available.

cell biology↗

bioRxiv2020

RNAseq analysis of rodent spaceflight experiments is confounded by sample collection techniques

To understand the physiological changes that occur in response to spaceflight, mice are transported to the International Space Station (ISS) and housed for variable periods of time before euthanasia on-orbit or return to Earth. Sample collection under such difficult conditions introduces confounding factors that need to be identified and addressed. We found large changes in the transcriptome of mouse tissues dissected and preserved on-orbit compared to tissues from mice euthanized on-orbit, preserved, and dissected after return to Earth. Changes due to preservation method eclipsed those between flight and ground samples making it difficult to identify spaceflight-specific changes. Follow-on experiments to interrogate the roles of euthanasia methods, tissue and carcass preservation protocols, and library preparation methods suggested that differences due to preservation protocols are exacerbated when coupled with polyA selection. This has important implications for the interpretation of existing datasets and the design of future experiments.

systems biology↗