NeTan is an end-to-end web platform that lets biologists turn raw metabolomics / transcriptomics (and other omics) tables into interactive networks and visual analytics — directly in the browser, no coding or local installs required.
| NeTan in a nutshell | |
|---|---|
| Upload anything | Mass-spec features, targeted panels, gene-expression counts, metadata — drag-and-drop CSV/TXT. Samples auto-align across files. |
| One-click pre-processing | Missing-value filter, quantile / median / mean normalisation, log₂ transform, Pareto or unit-variance scaling. |
| Smart feature filtering | Built-in t-test, one-way ANOVA, PLS-DA VIP, log-fold change, logistic / linear regression, RF-importance. |
| Network inference | Spearman, CLR (MI z-score), Random-Forest similarity, Graphical Lasso. Single-layer, stacked or true multilayer aggregation (mean / median / max). |
| Instant exploration | Force-directed / Kamada-Kawai / circular layouts, colour & shape nodes by any meta column, edge-weight slider, hide isolates, hover tooltips, legend toggles, full-screen. |
| Share & export | Download edge + node tables, PNG plots, or your processed data to reproduce the analysis elsewhere. |
- Open https://netan.io.
- Click “Upload Data” → drop your CSVs.
- Adjust parameters in the panel.
- Hit “Build Network” — watch the graph appear.
- Explore, filter, export. Done.
| Layer | Stack |
|---|---|
| Frontend | React 18, MUI v5, Plotly.js |
| Backend | Django 4, DRF, Pandas, SciPy, Scikit-learn, NetworkX |
| Realtime | WebSocket job status + progress bars |
| Deploy | Gunicorn + Nginx on AWS EC2 |
Boris Minasenko
📧 boris.minasenko@emory.edu