idw – Informationsdienst Wissenschaft
Nachrichten, Termine, Experten
Nachrichten, Termine, Experten
Faster, better, and ideally tailored to each individual: the medicine of the future aims to deliver precise diagnoses and treatment plans. The interaction of proteins plays a key role in this. To understand this interplay, large, high-quality datasets must be analyzed as efficiently as possible. MSAID, a spin-off of the Technical University of Munich (TUM), has developed AI-powered software that can do this even for complex samples.
Some ideas won’t let go. That was the case for a team at Professor Bernhard Küsters Chair of Proteomics and Bioanalytics at TUM. Building on a research project, the scientists developed a software prototype to determine which proteins are present in a sample, when and in what quantities they appear, and what they do. Gaining a better understanding of these processes could enable medical breakthroughs in diagnosing and treating disease.
“Our prototype already outperformed existing approaches in both the quality and quantity of proteins identified. We were determined to push this development forward and make it accessible to researchers and institutions worldwide,” says Mathias Wilhelm, Professor of Computational Mass Spectrometry at the TUM School of Life Sciences. Mathias Wilhelm and Bernhard Küster already had founded a company together and decided to take the leap again. Together with Martin Frejno, Daniel Zolg, Siegfried Gessulat, and Tobias Schmidt, they founded MSAID in 2019, a startup specializing in deep-learning models for protein research. All founders have studied, earned their doctorates, or worked at TUM.
A database for AI-generated molecular “fingerprints”
Since 2022, MSAID has been marketing a patented successor of the software whose prototype sparked the company’s founding. It streamlines the analysis of complex, large datasets. MSAID COO Daniel Zolg compares the approach to fingerprint identification: “Every protein consists of different peptides, each with its own specific fingerprint. We can make the fingerprints in a sample visible using a mass spectrometer. But the quality of these fingerprints isn’t always good—often you only have a partial print of a peptide, and sometimes they even overlap. That makes it harder to tell which measurement corresponds to which peptide. It’s a bit like trying to analyze fingerprints on a doorknob that’s been touched by many different people.”
Manually analyzing such large amounts of data is no longer possible. After all, humans are made up of more than 20,000 different proteins. The founders therefore leverage AI’s strengths in pattern simulation: an algorithm compares the measurement results with a kind of “peptide database” containing AI-generated patterns created by the team. These patterns match those produced by peptides during mass-spectrometry-based analysis. And more: they can even be used to predict what patterns will emerge when different peptides overlap.
“Using our approach, we can substantially improve protein identification in complex samples like tissue and plasma, better quantify their amounts, and reduce manual steps. Our software also enables analyses that would otherwise take several weeks to be completed in just a few days,” says MSAID CEO Martin Frejno. “That opens new avenues for early disease detection, personalized medicine, and drug development.”
Further information:
The innovation ecosystem with TUM at its center is considered one of the most successful deeptech hubs in Europe. Its particular strengths are its strong, diverse network and extremely specific support. In initiatives and co-labs, start-ups work on innovations with established companies, experts, investors and administration. TUM and UnternehmerTUM, the Center for Innovation and Entrepreneurship, support start-up teams with programs that are precisely tailored to the individual phases of the start-up and the teams. The TUM Venture Labs offer direct access to cutting-edge research, technical infrastructure and market expertise in twelve fields of technology. Most recently, more than 100 companies were founded at TUM in one year and more than 1,100 start-up teams were supported by UnternehmerTUM and the Venture Labs. UnternehmerTUM, which invests with its own venture capital fund, has twice been voted Europe's best start-up center by the Financial Times.
Merkmale dieser Pressemitteilung:
Journalisten, Wirtschaftsvertreter
Informationstechnik, Medizin
überregional
Forschungs- / Wissenstransfer
Englisch
Sie können Suchbegriffe mit und, oder und / oder nicht verknüpfen, z. B. Philo nicht logie.
Verknüpfungen können Sie mit Klammern voneinander trennen, z. B. (Philo nicht logie) oder (Psycho und logie).
Zusammenhängende Worte werden als Wortgruppe gesucht, wenn Sie sie in Anführungsstriche setzen, z. B. „Bundesrepublik Deutschland“.
Die Erweiterte Suche können Sie auch nutzen, ohne Suchbegriffe einzugeben. Sie orientiert sich dann an den Kriterien, die Sie ausgewählt haben (z. B. nach dem Land oder dem Sachgebiet).
Haben Sie in einer Kategorie kein Kriterium ausgewählt, wird die gesamte Kategorie durchsucht (z.B. alle Sachgebiete oder alle Länder).
