Science & Technology
At the core of our innovation is the CYTOCAST DIGITAL TWIN Platform™, a cutting-edge high-performance computing platform that leverages a particle-based stochastic simulation algorithm to replicate the intricate interactions of proteins within a virtual cell. By modeling the complexation, decomplexation, and diffusion of proteins within cells, their compartments, and membranes, our simulator delivers both qualitative and quantitative insights into the cellular complexome.
Our approach is grounded in the principle that most biological functions are carried out by protein complexes. The cell simulator enables us to predict patient responses to treatments by simulating protein complex formation across multiple tissues, providing valuable data for personalized medicine.
A key aspect of our platform is the integration of diverse drug data from publicly available databases into our proteome-wide simulation pipeline. By simulating drug perturbations, we incorporate multiomics data specific to both the drug and the targeted cell type. These simulations are analyzed statistically to identify significant changes in protein complex abundance and structure caused by perturbations. Importantly, these changes are correlated with potential off-target effects and side effects, making off-target prediction and safety assessment a central focus of our platform.
The CYTOCAST DIGITAL TWIN Platform™ – Workflow Overview
The diagram illustrates how the CYTOCAST'S DIGITAL TWIN Platform™ processes drug candidate information to generate actionable insights for customers.
1. Input: Drug Candidate Information
- The process begins with the customer providing drug candidate information, typically in the form of a SMILES code.
- Cytocast integrates this input with partner data, incorporating various biological datasets.
2. Data Integration & Analysis
The Cytocast platform processes the input by leveraging:
- Drug target identification – Determining which proteins the drug is expected to bind to.
- Proteomics data – Understanding protein expression levels and interactions.
- Protein interaction networks – Mapping how proteins interact with each other in different cellular environments.
- Complex formation pathways – Studying how protein complexes are assembled and perturbed by drug interactions.
3. Predictive Modeling with AI-Powered Tools
The Cytocast platform uses three core AI-driven components to analyze drug behavior:
- Cytocast Off-Target Predictor (AI-powered) – Predicts off-target interactions, identifying unintended protein bindings that may lead to adverse effects.
- CYTOCAST DIGITAL TWIN Cell™ – Simulates protein complex perturbations, modeling how the drug affects cellular environments at a molecular level.
- Cytocast (Side) Effect Predictor (AI-powered) – Predicts potential side effects and broader drug effects, helping researchers assess safety risks.
4. Cytocast Report as an Output
The platform generates a comprehensive interactive report, which is delivered to the customer. This report includes:
- Predicted off-target interactions – Identifying unintended binding sites.
- Protein complex perturbations – Showing how the drug alters molecular networks.
- Predicted effects and side effects – Highlighting potential risks for further investigation.
Why It Matters
By leveraging Cytocast’s AI-powered predictive modeling, researchers and pharmaceutical companies can evaluate drug safety and efficacy early—before investing in costly experiments or clinical trials. This accelerates drug discovery, reduces development risks, and helps refine molecular designs for safer and more effective therapeutics.
Interested in a demo?
Contact us today to learn how the CYTOCAST DIGITAL TWIN Platform™ can revolutionize your R&D efforts and bring you closer to breakthroughs in drug safety and efficacy.
Contact usResearch Projects
As a research-driven company, Cytocast is deeply involved in a variety of collaborative projects that demonstrate the adaptability and impact of our technology. Tailored to address specific research challenges, these projects showcase how our solutions can advance understanding across a range of scientific fields.
- Modeling Protein Abundance's Impact on Postsynaptic Protein Network Organization
In collaboration with the Faculty of Information Technology and Bionics at Pázmány Péter Catholic University, we are investigating the formation of postsynaptic protein complexes. The goal is to understand the diversity of synaptic protein complexes within the postsynaptic density (PSD), a dense protein network essential for learning, memory, and neurological health. PSD composition is highly dynamic, varying across brain regions, developmental stages, circadian rhythms, andin response to external stimuli, making its molecular organization challenging to characterize.
Our project employs large-scale stochastic simulations of protein binding events to predict the formation and distribution of PSD complexes. Using experimental protein abundance data from various brain regions of patients, we modeled seven key PSD proteins. Results reveal that subtle changes in protein abundance can significantly influence the ratios of protein ícomplexes. These simulations are essential for understanding how protein availability drives complex formation and how these processes impact neurological function and disorders.
Stay tuned for updates on this and other exciting research projects that highlight the versatility of our platform. (Read more here and here)
- Unveiling COVID-19’s Neurological Impact with Cytocast
Recent research leveraging Cytocast has uncovered how COVID-19 disrupts brain function through microglial dysfunction, vascular inflammation, and blood-brain barrier damage. By simulating protein complex changes and mapping key pathways like VEGF and the IL-1/IL-6 axis, Cytocast reveals critical insights into SARS-CoV-2’s neurological effects. These discoveries pave the way for targeted therapies to mitigate long-term brain damage and inflammation caused by the virus. Explore how Cytocast is transforming our understanding of complex diseases.
Further information and updates on other ongoing projects will be provided later.
Publications
Below, you can find the scientific papers written by Cytocast scientists introducing the fundamental science behind our product. These papers illustrate how our research has contributed to the advancement of scientific knowledge in the fields of biology, bioinformatics, and health sciences.
List of publications:
- R. Fekete, A. Simats, E. Bíró, B. Pósfai, Cs. Cserép, A. D. Schwarcz, E. Szabadits, Zs. Környei, K. Tóth, E. Fichó, J. Szalma, et al. 'Microglia dysfunction, neurovascular inflammation and focal neuropathologies are linked to IL-1- and IL-6-related systemic inflammation in COVID-19', Nature Neuroscience, 28 : 3 pp. 558-576. , 19 p., (2025)
- A. Csikász-Nagy, E. Fichó, S. Noto, and I. Reguly, ‘Computational tools to predict context-specific protein complexes’, Current Opinion in Structural Biology, vol. 88. Elsevier BV, p. 102883, Oct. 2024.
- M. Miski, Á. Weber, K. Fekete-Molnár, B. M. Keömley-Horváth, A. Csikász-Nagy, and Z. Gáspári, ‘Simulated complexes formed from a set of postsynaptic proteins suggest a localised effect of a hypomorphic Shank mutation’, BMC Neuroscience, vol. 25, no. 1. Springer Science and Business Media LLC, Jul. 06, 2024
- Miski, Marcell, Bence Márk Keömley-Horváth, Dorina Rákóczi Megyeriné, Attila Csikász-Nagy, and Zoltán Gáspári. "Diversity of synaptic protein complexes as a function of the abundance of their constituent proteins: A modeling approach." PLOS Computational Biology 18, no. 1 (2022)
- Quaglia, Federica, Bálint Mészáros, Edoardo Salladini, András Hatos, Rita Pancsa, Lucía B. Chemes, Mátyás Pajkos et al. "DisProt in 2022: improved quality and accessibility of protein intrinsic disorder annotation." Nucleic Acids Research 50, no. D1 (2022): D480-D487.
- Pancsa, Rita, Erzsébet Fichó, Dániel Molnár, Éva Viola Surányi, Tamás Trombitás, Dóra Füzesi, Hanna Lóczi et al. "dNTPpoolDB: a manually curated database of experimentally determined dNTP pools and pool changes in biological samples." Nucleic Acids Research 50, no. D1 (2022)
- Reguly, István Z., Dávid Csercsik, János Juhász, Kálmán Tornai, Zsófia Bujtár, Gergely Horváth, Bence Keömley-Horváth et al. "Microsimulation based quantitative analysis of COVID-19 management strategies." PLoS computational biology 18, no. 1 (2022)
- Rizzetto, Simone, and Attila Csikász-Nagy. "Toward large-scale computational prediction of protein complexes." Computational Cell Biology: Methods and Protocols (2018): 271-295.
- Rizzetto, Simone, Petros Moyseos, Bianca Baldacci, Corrado Priami, and Attila Csikász-Nagy. "Context-dependent prediction of protein complexes by SiComPre." NPJ Systems Biology and Applications 4, no. 1 (2018): 37.
- Rizzetto, Simone, Corrado Priami, and Attila Csikász-Nagy. "Qualitative and quantitative protein complex prediction through proteome-wide simulations." PLOS Computational Biology 11, no. 10 (2015): e1004424.