According to research published in the journal Nature, a new search platform called Malva enables scientists to search RNA sequences across millions of individual cells in seconds without downloading enormous raw datasets or mapping them against a reference genome. Developed by creators including Nikolaus Rajewsky, Daniel León-Periñán, and Nikos Karaiskos, the system has indexed more than 140 terabytes of public single-cell and spatial transcriptomics data to help researchers investigate cancer, infections, and other biological conditions.
How Malva Indexes Massive Single-Cell Datasets Without Reference Genomes
Single-cell RNA sequencing generates petabytes of sequence information from hundreds of millions of cells, creating a massive data management challenge for modern biology laboratories. Traditional portals organize this information around predefined genes by mapping sequencing reads to a reference genome and summarizing them into gene-level counts. While effective for basic questions, this approach often buries original sequence-level details. According to the creators, Malva bypasses the need to download petabytes of raw files by dividing sequencing reads into short segments called k-mers and recording which cells contain them. The human index described in the study contains more than 100 billion unique 24-nucleotide sequences from roughly 51 million cells representing 592 studies and 7,966 samples, alongside about 10 million mouse cells.
Sequence-Based Queries for Mutations and RNA Isoforms
Instead of restricting queries to gene activity counts, Malva accepts nucleotide sequences, gene identifiers, or natural-language requests to return matching cells and sample information. According to Nikos Karaiskos, the platform allows researchers to explore complex features like splice junctions, RNA isoforms, viral sequences, and bacterial or fungal material that conventional gene-count databases miss. In performance tests, a single k-mer search took about 70 milliseconds, a 1,000-base transcript search took about 0.9 seconds, and searching 1,000 transcripts took roughly one minute on a single CPU core. The system successfully detected somatic mutations across 16 cancer types, identified viral and laboratory contaminants like Mycoplasma, and recovered cell-specific patterns involving untranslated regions of RNA without requiring specialized mutation-calling software.
Did you know? Malva can also process spatial transcriptomics data, allowing researchers to determine precisely where particular RNA sequences appear within tissue sections, such as tracking the circular RNA CDR1as to find that 95% of positive cells are excitatory neurons.
Bridging Experimental Cellular Evidence with Artificial Intelligence
As artificial intelligence becomes more prevalent in life sciences research, systems like Malva offer a direct way for AI tools to query underlying experimental cellular evidence. According to Nikolaus Rajewsky, Malva transforms static transcriptomic atlases into dynamic resources that can clarify how health shifts into disease and how specific cells respond to medical treatments. Despite these capabilities, the platform has limitations: it currently requires queries of at least 24 nucleotides, relies on exact sequence matching, and reports k-mer-derived pseudocounts rather than absolute molecule counts. Academic scientists can currently access Malva through its public interface and API, while the creators pursue a patent and early-stage commercialization efforts.
Frequently Asked Questions
What is Malva?
Malva is an RNA sequence search platform that lets scientists search millions of individual cells in seconds without downloading enormous raw datasets or using a reference genome, according to research published in Nature.
How much data has Malva indexed?
The Malva Index has processed more than 140 terabytes of publicly available single-cell and spatial transcriptomics data, encompassing over 100 billion unique 24-nucleotide sequences from roughly 51 million human cells and 10 million mouse cells.
Who created Malva?
The platform was developed by a research team including Nikolaus Rajewsky, Daniel León-Periñán, and Nikos Karaiskos.
Can Malva detect cancer mutations?
Yes. Testing showed that Malva could detect somatic mutations across 16 cancer types without requiring researchers to realign underlying sequencing reads or run specialized software.
Call to Action: Explore our latest coverage on single-cell genomics and spatial transcriptomics, and share your thoughts in the comments below on how high-speed sequence indexing will change bioinformatics research.