Interactive data exploration
Explore gene-expression matrices, other biological measurements and general feature-and-sample datasets through clustering, projections and linked visualizations.
RESONALIS / EXPLORING WHAT COMES NEXT
Beyond data.
Into understanding.
Exploratory analysis for biological data and modern machine learning — connecting multivariate statistics, interactive visualization and the emerging capabilities of artificial intelligence.
Rooted in bioinformatics. Expanding into AI. One shared challenge: finding structure in complex data.
01 / OUR VISION
Make complexity explorable.
The most valuable patterns are rarely obvious. We believe researchers and engineers need more than automated answers: they need the ability to explore, question and understand their data. MolMine is building toward a workspace where human insight and intelligent computational tools complement one another.
02 / WHY J-EXPRESS RETURNS
Different data.
Shared methods.
When J-Express first emerged, researchers faced thousands of gene-expression measurements and needed ways to discover relationships hidden in high-dimensional data. Clustering, principal component analysis, self-organizing maps and linked visualization helped turn those measurements into explorable patterns.
Biological research still needs these methods, while machine learning and foundation models generate high-dimensional embeddings and learned representations. Across both fields, many fundamental questions converge: Which observations belong together? What separates them? Where are the outliers? Which features explain a pattern?
That is why we are bringing J-Express back — not to leave bioinformatics behind, but to extend the same exploratory methods to a new field. The workspace supports biological feature-and-sample matrices and other compatible high-dimensional numerical datasets. Understanding the mechanisms of AI models will also require new, model-aware methods.
One analytical foundation. From biological measurements to AI representations.
03 / THE APPROACH
Classical insight.
Modern intelligence.
Our approach connects biological data analysis and machine-learning exploration through shared methods: interpretable statistics, clustering, projections and interactive inspection.
Machine learning & embeddings
Explore model-generated text embeddings using PCA, UMAP, cosine nearest neighbors, semantic search and interactive neighborhood analysis.
AI-assisted discovery
We aim to give AI agents access to explicit, reproducible analysis operations — allowing them to suggest hypotheses, run tools and help interpret results.
04 / MODERN RNA-SEQ ANALYSIS
From expression counts
to biological insight.
Modern RNA sequencing produces expression measurements for thousands of genes across samples. J-Express helps you explore the resulting gene-by-sample matrix: inspect sample structure with PCA and correspondence analysis, compare expression patterns using heatmaps and hierarchical clustering, investigate individual gene profiles, and use feature selection to identify informative genes.
Import a prepared, appropriately normalized or transformed expression matrix as CSV/TSV, with genes as features and samples as columns. Sample metadata and groups help interpret biological contrasts. J-Express is an interactive downstream exploration environment, not a replacement for read alignment, transcript quantification or RNA-seq-specific count modeling such as DESeq2/edgeR. Do not interpret general-purpose differential-analysis results as validated RNA-seq count statistics.
Try the biological demo
Open J-Express, choose Demo → Gene expression (export any unsaved project first), then select the demo dataset in Project Explorer. Start with Views → PCA to inspect sample relationships; open Measurement Heatmap or Hierarchical clustering to compare genes and samples; then try Measurement Profiles and Feature Subset Selection (FSS) to inspect or rank expression patterns. Group colors and linked selection help follow the same samples/features between views.
The bundled Gene expression demo comes from a legacy J-Express .pro project; it is not an RNA-seq sequencing/count dataset. It demonstrates the exploration workflow. For a real RNA-seq study, first prepare an appropriate expression matrix with sample annotations in an RNA-seq analysis pipeline.
04 / AI INTERPRETABILITY
Beyond predictions.
Toward understanding.
AI models are becoming more capable, but understanding what happens inside them remains a major research challenge. Why does a model respond differently to similar inputs? Which internal representations change when unexpected behavior appears?
Resonalis aims to bring the exploratory strengths of J-Express to this problem: analyzing activation patterns, comparing groups of model runs and visualizing high-dimensional representations. These analyses may help researchers identify candidate features and formulate testable hypotheses about model behavior.
Finding correlations is not the same as establishing causes. Explaining why a model behaves a certain way also calls for model-aware experiments and causal intervention methods — directions for future research, rather than capabilities we claim to provide today.
Make the hidden structure of AI models easier to explore.
05 / HOW IT COULD WORK
From a question to a discovery.
J-Express already lets researchers inspect multidimensional data through linked visualizations. We are extending these workflows to modern embeddings, while automated AI-agent analysis remains a research direction.
Bring your data
Import biological datasets, existing J-Express projects, numeric CSV/TSV matrices or embedding vectors. J-Express also includes real text-embedding demonstrations and optional browser-local MiniLM inference.
Follow the patterns
Use clustering, projections and linked visualizations to explore structure, outliers and meaningful differences.
Work with an AI agent
Ask questions, evaluate proposed analyses and examine the evidence behind each finding. Agent-driven workflows are part of our roadmap.
Illustrative future workflow — not a claim that all AI integrations are available today.
06 / THE PLATFORM
J-Express.
Reimagined.
J-Express is back as a browser-based data exploration workspace. For biological data it supports PCA, correspondence analysis, hierarchical clustering and heatmaps, self-organizing maps, measurement profiles, feature selection, and differential analysis including SAM where applicable. Legacy J-Express .pro projects can be imported, and projects can be saved locally or exported as JSON.
J-Express also supports embedding datasets, UMAP, cosine nearest neighbors, semantic search, Concept Explorer and neighborhood-preservation analysis. A MiniLM text demo shows how sentences become numerical vectors; compatible new text can be embedded locally in the browser when model assets are available. These techniques help reveal relationships in representations; they do not, by themselves, explain the causal mechanisms of an AI model.
Biological analysis remains central. Direct agent-controlled workflows and broader model integrations are planned rather than available features.
Try J-Express07 / SEE IT IN ACTION
Explore J-Express today.
Explore biological data and AI embeddings in the same browser-based J-Express workspace, from clustering and interactive feature analysis to UMAP, semantic search, and exploration of model-generated representations.
J-Express is evolving. Generating new text embeddings can require downloading a model after consent. Neither projection nor semantic similarity establishes causality.