AstroOmix
AstroOmix is a Claude-powered research application for analysing space-biology data without overstating what limited datasets can support.
Core trick
Deterministic differential-expression and longitudinal models compute statistics, and every number in the Claude answer is checked against tool output before display.
Description
AstroOmix is a Claude-powered research application for analysing space-biology data without overstating what limited datasets can support.
Researchers can upload their own data or explore real NASA studies through two workflows: differential-expression analysis for replicated spaceflight RNA-seq experiments, and longitudinal modelling of physiological measurements across pre-flight, flight and recovery phases. Deterministic tools perform the statistics, while an embedded Claude agent explains results, answers study-specific questions and retrieves relevant PubMed evidence.
AstroOmix is designed around verifiability. Numerical claims are checked at runtime against structured analysis outputs; unsupported figures are withheld. Literature citations must originate from retrieved records, and literature consistency is presented as context, not proof. The application also flags thin time series, distinguishes interpolation from extrapolation and avoids clinical conclusions.
Building AstroOmix highlighted a central challenge in space-health research: the available data are valuable, but small, heterogeneous and easy to overinterpret. The project therefore focuses not on automating certainty, but on making analysis traceable, reproducible and appropriately cautious.
As longer missions generate increasingly complex biological data, tools that combine computational analysis with evidence-aware AI reasoning can help researchers reach useful conclusions while remaining explicit about uncertainty.
README excerpt
# AstroOmix **Live demo:** [astroomix.streamlit.app](https://astroomix.streamlit.app/) AstroOmix is a research web app for analysing space-biology data. It runs two workflows over real NASA datasets — bulk RNA-seq differential expression (rodent spaceflight vs. matched ground control) and longitudinal modelling of human biosignals across a mission — and pairs them with an AI research assistant that is constrained to the computed results. The assistant has no independent knowledge of these datasets: it calls the same analysis code the pages run, every tool call it makes is shown to you, and **every figure in its answers is verified against the actual tool output before you see it.** Anything that cannot be traced is withheld rather than shown. It also retrieves real published papers from PubMed — retrieval only, displayed with their abstracts for you to judge, never presented as validation of a result. Space-biology datasets are small, heterogeneous, and easy to overinterpret. The point of this project is not to squeeze more confidence out of them than they can support; it is to make the limits of the data visible while the analysis is being read. ## What it does | Workflow | What it is | |---|---| | Differential expression | Flight vs. ground control on bulk RNA-seq. PyDESeq2 for the worked examples; a resource-safe NB-GLM for uploaded data. Benjamini–Hochberg FDR. | | Longitudinal analysis | Prophet / ARIMA / LightGBM compared by leave-one-out CV on a repeated-measures