Foundation models
Organize unstructured research and support real-time health decisions during long-duration missions.
NASA Ames Research Center
Computational Space Biology Lab
Turning space biology data into insight, so crews can travel farther and live healthier on Earth and beyond.
SPACE BIOSCIENCES / RESEARCH BRANCH
Earth, aurora, and stars from the International Space StationNASA/JSC · Expedition 43
The Computational Space Biology Lab builds computational frameworks that use machine learning and artificial intelligence to model, predict, and help mitigate the risks of spaceflight.
Connecting biological data, computational methods, and human space exploration.
Research directions
Questions worth
solving in space
We develop tools to make complex biology more understandable, predictive, and useful for the people planning future missions.
Digital twins
Model how biological systems respond to spaceflight stressors and explore precision interventions.
Causal inference
Identify relationships in complex biological systems to improve prediction and risk mitigation.
Multi-modal data integration
Bring diverse NASA Open Science Data Repository datasets together for richer analysis.
Federated learning
Train models across separated datasets while reducing data movement and respecting privacy.
Knowledge graphs
Connect biological evidence across data sources to reveal patterns relevant to crew health.
From data to discovery
Current projects
Methods, datasets, and collaborations advancing AI/ML for space biology.
Explainable AI · Multi-omics
Reading the signals in muscle
Graph-based explainable AI connects gene expression, proteins, methylation, and calcium uptake in spaceflown mouse muscle.
Read the studyCausal machine learning
Finding what drives change
The CRISP ensemble identifies genes potentially causal of increased liver fat in spaceflown mice, with work expanding to bio-imagery.
Read the studyTransfer learning
Learning from small datasets
Guided Transfer Learning uses inductive biases to help models learn effectively in high-dimensional, low-sample-size space biology.
Explore the preprintDistributed computing
Learning across Earth and orbit
A federated approach trains predictive models across physically separated nodes, including data on the International Space Station.
Read the preprintKnowledge networks
Connecting biology to health
NASA GeneLab and the SPOKE biomedical knowledge graph help link spaceflight biology to physiological changes and symptoms.
Read the studyResearch community
Horizons in Biosciences & Informatics
A seminar series bringing researchers together to discuss emerging methods, data, and interdisciplinary discoveries.
Explore the seriesGenerative AI · Data science
Building realistic synthetic data
Generative adversarial networks amplify signal in small space biology datasets using synthetic RNA sequencing data.
Read the paperOpen science · Benchmarking
RNA sequencing benchmark
A standardized, AI-ready dataset for model benchmarking, training, and community data challenges.
Access the datasetOpen science · Benchmarking
Microscopy benchmark
A standardized fluorescence microscopy dataset, supported by NASA, for AI/ML benchmarking and research challenges.
Access the datasetCommunity · Education
AI/ML space biology course
Open-source, no-cost virtual training in AI/ML for space biology, designed with no prerequisites.
Enroll in the coursePeople behind the research
Research across disciplines
The Computational Space Biology Lab brings together expertise in machine learning, biology, biomedical research, and spaceflight. Connect with the team or join the open science community.
Co-leads
2 researchers
Current members4 researchers
Christina JohnsonVishnu Prasad
Amanda Saravia-ButlerJatin Chaudhary
Former team members18 researchers
Konstantinos AdamopoulosSarah GoltsKevin LiPhani Paladugu
Sylvain CostesAdrienne HoarfrostMakenna MyrickSonali Patel
Kayvon CrenshawLiam JohnsonCharlotte NelsonLauren Sanders
Riya DesaiAlice JoyeuxDillon NishigayaAnmol SethRyan Scott
Brian Evarts
Selected work
Research in print
Recent publications and preprints from Computational Space Biology Lab collaborators.
Machine Learning Ensemble Investigates Age in the Transcriptomic Response to Spaceflight in Murine Mammary Tissue: Observational Study. JMIRx Bio.
↗Bridging Earth and Space: A Flexible and Resilient Federated Learning Framework Deployed on the International Space Station. bioRxiv.
↗Analyzing the relationship between gene expression and phenotype in space-flown mice using a causal inference machine learning ensemble. Scientific Reports.
↗Explainable machine learning identifies multi-omics signatures of muscle response to spaceflight in mice. npj Microgravity.
↗