svarogjk/caloric_restriction: The first version for zenodo
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
GEO Survival Analysis v0.1.0 Cross-dataset biomarker discovery from NCBI GEO — no coding required. What it does Type a natural language query (e.g. "lung adenocarcinoma overall survival") and get a ranked, cross-validated list of survival-associated genes backed by multiple independent GEO cohorts — complete with Kaplan-Meier curves, forest plots, hazard ratios, and confidence intervals. Features in this release Cross-cohort meta-analysis — ranks genes by consistency across independent datasets, not just one study Natural language input — no GEO accession numbers or coding required Kaplan-Meier curves with CI bands, per-dataset and combined views Forest plots with per-dataset HRs, pooled effect, and I² heterogeneity Volcano plot — interactive HR vs. significance scatter Real-time progress via SSE streaming Gene batch mode — restrict analysis to up to 500 candidate genes COSMIC cancer gene filter — ~600 well-characterised cancer driver genes CSV + PNG export for all plots and tables Shareable permalinks — /results/:id renders publicly without login Publication export — ZIP with genes.csv, per-dataset results, and a pre-written Methods paragraph AI chat assistant with RAG context over analyzed datasets Analysis history and comparison dashboard Stack FastAPI · PostgreSQL + pgvector · lifelines · LangChain 1.x · React 18 · Redux Toolkit Cite this software Korkodnov I. et al. GEO Survival Analysis: cross-cohort biomarker discovery from NCBI GEO.
Fields of Study & Subjects
Gene expression and cancer classification
Bioinformatics and Genomic Networks
Genetics
Aging
and Longevity in Model Organisms
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 28 March 2026
- Language
- EN