Laboratory seminars (PV273 in the course catalog) on Wednesday, 10:30 – 11:30, A505, FI MU, Botanická 68, followed by an informal group lunch.

The format of standard lectures: 30-40 minutes presentation + 15 minutes for questions. Slides are expected in English, and the presentation is in English or Czech, depending on the audience. For past seminars, go here.

• 16.9.2026
  Estrella Fernandez-Gimenez, PhD
  SBGrid: Research Computing Support for Structural Biology
  Abstract: Modern structural biology depends on complex software ecosystems that are difficult to install, maintain, and reproduce, particularly for cryo-EM. The non-profit SBGrid Consortium supports over 500 laboratories in 22 countries by centrally maintaining and distributing more than 1,000 applications. SBGrid promotes sustainability and reproducibility through software “capsules” that bundle applications with their dependencies for consistent cross-platform use. Through the Instruct-SP partnership and collaboration with the JM Carazo laboratory, SBGrid has integrated Scipion plugins directly with its software capsules. This integration provides reproducible cryo-EM workflows, shared environments, coordinated training, and local support. SBGrid also has a cloud platform that enables computationally intensive workflows without extensive local infrastructure. It offers a proven model for scalable and sustainable structural biology computing across Europe.
• 23.9.2026
  RNDr. David Procházka, Mgr. Filip Jozefov, Mgr. Emma Sommerová, Bc. Boris Lukačovič
  DreaMS Search: vector search over 200+ million tandem mass spectra
  Abstract: Small molecules are central to biology and are an important source of medicines and other bioactive compounds. Yet the molecular structures of most small molecules observed in biological and environmental samples remain unknown. Tandem mass spectrometry is one of the main technologies used to study these molecules. Vast amounts of mass spectrometry data have been collected and shared through GNPS/MassIVE, a major public repository that now contains more than a billion spectra, most of them without an identified molecular structure. The recently introduced model DreaMS (Bushuiev et al., Nature Biotechnology, 2026) represents spectra so that their similarity reflects the chemical similarity of the molecules that produced them.
  To make the power of DreaMS available to the community, we introduce DreaMS Search (dreams.fi.muni.cz), a large-scale similarity-search application that currently provides access to more than 200 million MS/MS spectra from GNPS/MassIVE. In this presentation, we describe the application: its architecture, the spectral data and metadata it indexes, the interface through which chemists explore search results, and how DreaMS is being extended beyond spectral search.
  DreaMS Search is a joint effort of FI MUNI, IOCB Prague, CIIRC CTU, CESNET, and Seoul National University.
• 30.9.2026
  RNDr. Pavel Novák
  Early malware detection from event sequences: From behavioral analysis to real-time decision-making
  Abstract: Traditional malware detection often relies on known indicators of compromise or evaluates program behavior only after a sufficient amount of information has been collected. From the perspective of active system protection, however, a more important question arises: Can malicious behavior be recognized early enough, before the malware carries out its key activities?
  This talk presents doctoral research focused on the early detection of malware using sequences of events that capture its behavior during execution. The core idea is not to evaluate individual events in isolation, but to continuously assess whether the behavior observed so far is more consistent with a benign or a malicious program. This results in a continuously updated decision process that may enable detection based on only a partial execution trace.
  The talk will introduce the main principles of the proposed approach, the representation of program behavior, and its experimental evaluation using data obtained from dynamic malware analysis.
  The results show that characteristic differences between benign and malicious behavior can already be observed at relatively early stages of program execution, and that strong detection performance does not necessarily require highly complex models.
• 7.10.2026
  Raheleh Abbasi
  A newcomer’s perspective: Relocation, academic life, and my PhD journey in Czechia
  Abstract: Moving to a new country involves much more than adapting to a new place; it also means becoming familiar with a different culture, academic environment, and way of everyday life. In this talk, I will share some of my personal experiences of relocating to Czechia and becoming part of the university and research environment as an international PhD student. I will reflect on aspects of the Czech academic and educational system that I found interesting, different, or unexpected from an external perspective, as well as the practical and cultural challenges of settling into a new country. I will also highlight aspects that I have particularly appreciated and share a few observations that may be useful when welcoming and supporting international students and researchers. Finally, I will briefly introduce my PhD project, its main goals and research direction, and my plans for the next stages of my PhD.

Contact: Hana Rudová