See Malva in Action
Experience the power of genomic search
Results Summary
Real-time Search across Millions of Cells
Instantly search across petabytes of single-cell genomics data, enabling rapid sequence, gene and pathway quantification across millions of cells.
Unlocking Omics Data on Four Levels
Malva introduces a paradigm shift in data processing, enabling target identification, patient stratification, and individualized understanding of disease mechanisms.
Novel Indexing Algorithm
A unified, searchable database of omics data, enabling immediate sequence discovery without preprocessing.
Ultra-fast Search Engine
Instant searches in petabytes of data using diverse query possibilities, revealing biological mechanisms.
Intuitive Visualization
User interface facilitates seamless exploration and interpretation of complex datasets.
AI-powered Exploration
Integrates with Large Language Models to enable natural language queries and automated discovery.
Choose the Access Type for Your Research
Unlock the full potential of single-cell and spatial genomics with Malva.
Open
Access to public datasets for educational and research purposes.
- Public datasets (2024)
- Login with ORCID account
- Premium indices
- AI-assisted analysis
Individual
Full access to all indexed datasets and advanced features.
- Everything in Open
- Premium indices
- Custom accounts
- AI-assisted analysis
Institutional
Tailored solutions for organizations with custom requirements.
- Everything in Individual
- Custom dataset upload
- Private deployments
- SLA Support
Frequently Asked Questions
Learn more about Malva and our team
Malva is a universal search index for single-cell and spatial sequencing technologies that unifies multi-terabyte repositories of nucleotide data into an instantly queryable resource. The system allows researchers to search across hundreds of millions of cells from thousands of experiments with single-cell and spatial resolution.
Malva uses advanced indexing technology to store and query cellular transcriptomic data efficiently. The core components include a k-mer based indexing system for ultra-fast sequence search, distributed storage architecture for scaling to billions of cells, and natural language processing to convert plain English queries into gene and sequence searches.
Malva supports three primary query types: DNA/RNA sequences (e.g., "ATGCTAGCTAGCT"), Gene IDs (e.g., "CD28", "IL2RA"), and Natural Language questions (e.g., "Find T cells expressing CD28 in lung tissue").
Malva currently indexes over 3,500 datasets from major single-cell and spatial transcriptomics initiatives, including Human Cell Atlas (HCA), Public GEO-archived spatial samples, Tabula Muris, Linnarsson Adult Human Brain, and numerous tissue-specific atlases.
If you use Malva in your research, please cite our paper: León-Periñán D, Karaiskos N, Rajewsky N (2025) "Malva: A universal search index for single-cell and spatial transcriptomics."