Summary
- Profile Type
- Technology offer
- POD Reference
- TOHU20240910012
- Term of Validity
- 10 September 2024 - 10 September 2025
- Company's Country
- Hungary
- Type of partnership
- Research and development cooperation agreement
- Targeted Countries
- All countries
Contact the EEN partner nearest to you for more information.
Find my local partner
General information
- Short Summary
- A Hungarian techbio startup, specializing in cell simulation for drug discovery and using information on cellular processes, can predict protein complex changes in response to perturbations in diseases or drug treatments. The SME is looking for research partners with biotech or pharma companies involved in drug discovery to validate its platform. During a pilot study, the SME will test drug candidate(s), and provide valuable insights on related drug/side effects predicted on various cell types.
- Full Description
-
A Hungarian techbio startup, specializing in cell simulation for drug discovery and using information on cellular processes, predicts protein complex changes in response to perturbations, such as diseases and drug treatments. With expertise in computational biology, bioinformatics and high-performance computing, it builds a healthcare-oriented digital twin of patients using their multi-omics data, which allows rapid in-silico drug testing and enables personalized treatments.
Currently, taking a new drug to market takes several years, and hundreds of millions of dollars. Pharmaceutical companies struggle to predict drug side effects accurately during early development due to heterogeneity of patient responses and the complexity of biological systems. Adverse reactions, drug inefficacy, and financial losses occur, leading to concerns about healthcare and the withdrawal of drugs. Traditional trials are resource-intensive and fail to foresee all potential adverse reactions, necessitating a solution to optimize drug selection and reduce side effects.
The SME develops a high-performance computing tool that utilizes an agent-based stochastic simulation algorithm to create virtual cell models. It focuses on the interactions of proteins within the cell, allowing for qualitative and quantitative insights into the complexome of the simulated cell.
The process involves four main steps: (i) model generation, (ii) AI- based drug target and off-target prediction, (iii) simulation of the model, and (iv) ML supported reporting of potential side effects caused by drugs or drug combinations. It generates individual cell models from various tissues by integrating molecular data, supports drug testing with pre-defined drugs or custom drug-protein interaction data, and uses a parallelized simulation algorithm to evaluate the effects of each drug on requested tissue types. The resulting reports identify the most affected protein complexes and interactions, as well as the cellular phenotype changes caused by the drug in each tissue type.
The technology is based on a comprehensive cell simulator that integrates bioinformatics databases into a computer model to understand the impact of drug-related perturbations and predict how drugs can modify cellular health on multiple tissue types and cell lines.
The key user groups of the tool are the pharmaceutical industry, drug developers, and researchers involved in drug discovery and development. The technology can predict the effects and side effects of drugs during the R&D phase. By simulating changes in protein complexes and evaluating drug-protein interactions across multiple tissues, the software developed enables researchers to make informed decisions about drug candidates before starting clinical trials. Its predictive capabilities help pharma companies optimize drug selection, decrease side effects, enhance treatment efficacy, ultimately leading to more efficient drug development processes and increased success rates in clinical trials.
The SME has recently established a Pilot Study Program, and already signed contracts with three partners. The program aims to find partnerships with biotech and pharma companies actively involved in drug discovery. Partners are expected to validate the effectiveness of the technology, access real-world data, and gain support for its integration into drug development processes. The study will anticipatedly tackle some of the most pressing issues in drug development. Partners are expected to present with a challenging task – regardless of whether they have a drug candidate that faced setbacks in a particular phase or a set of candidates with intriguing side effects or drug-drug interactions, while they will be provided with valuable insights and solutions.
The SME is looking for R&D partners to build long term partnerships. - Advantages and Innovations
-
The SME is the first provider of predictive whole-cell level simulations of protein interactions.
Differently from many machine-learning and AI approaches, the software platform developed does not require any training dataset and it can provide information for drug development just from a single input. This makes such platform the only choice for diseases, where it is infeasible to generate large training datasets, e.g. in case of rare or orphan diseases and personalized treatments. Furthermore, the approach gives a methodological explanation of how diseases and drugs work, not just an answer with a statistical significance. Still, AI and machine learning algorithms are used in data pre- and post-processing to improve performance.
Even though the solution is general, applicable to most diseases, the first collaborators’ main focuses are neurodegenerative disorders, rare diseases, and immune-related applications for cancer research, which are often ignored by computational biology and bioinformatics companies because of the lack of data.
The platform developed is based on a unique approach that integrates data into a simulator of protein complex formation to understand physiological responses of cells and learns side effects associated with physiological changes. Other technologies train ML algorithms on diseases with wide knowledge to predict treatment outcome from mutations only. The simulator used in the platform applies all available molecular data and predicts protein complexes, i.e. the machineries of cells are coming together in each individual cell. It is far beyond competing approaches, where they train machine learning algorithms on diseases with wide knowledge to predict treatment outcome from mutations only. - Stage of Development
- Available for demonstration
- Sustainable Development Goals
- Goal 3: Good Health and Well-being
- IPR status
- Secret know-how
Partner Sought
- Expected Role of a Partner
-
- Type of partner sought:
The Pilot Study Program is open for industrial and academic partners as well as research institutions.
- Specific area of activity of the partner:
The key user group of the software tool is the pharmaceutical industry, drug developers, and researchers involved in drug discovery and development. The technology provides a powerful software tool capable of predicting the effects and side effects of drugs during the R&D phase. By simulating changes in protein complexes and evaluating drug-protein interactions across multiple tissues, the technology enables researchers to make informed decisions about drug candidates before starting clinical trials. Its predictive capabilities help pharmaceutical companies optimize drug selection, decrease side effects, and enhance treatment efficacy, ultimately leading to more efficient drug development processes and increased success rates in clinical trials.
- Task to be performed:
The partners involved in drug discovery are expected to validate the effectiveness of the software platform developed. The pilot study will test drug candidate(s) and provide valuable insights on related drug / side effects predicted on various cell types.
Key solutions of the platform used is aimed to solve:
• Early-Stage Candidate Testing: Cost-effective insights into potential side effects before committing to clinical trials.
• Trial Risk Mitigation: Identifying candidates with higher failure risk due to side effects and optimizing trial design.
• Animal to Human Transition: Evaluating candidates for successful transition, reducing risks in human trials.
The pilot study will anticipatedly tackle some of the most pressing issues in drug development. Partners are expected to present with a challenging task - regardless of whether they have a drug candidate that faced setbacks in a particular phase or a set of candidates with intriguing side effects or drug-drug interactions, while they will be provided with valuable insights and solutions.
The techbio startup offers to the partners to
• reduce the risk of failure in clinical trials by selecting best drug target from pre-clinical trial data, and
• improve your understanding of how patients respond to therapies – desired and non-desired effects of treatments.
• perform proteome-wide simulations of protein complexes.
As part of a pilot study, the benefits delivered to the partners are as follows:
• digital evaluation of the partner’s drug candidates before a clinical trial.
• a unique insight into drug-effect mechanisms.
• assistance to find new applications for partner’s drugs that’s already approved through digitalized drug repositioning.
• fine-tuned dosage recommendations as well as test results on multiple tissue types to assess system-wide effects and side-effects when applying a single drug or multiple drugs simultaneously.
The SME expects from the partners:
• Input data (SMILES code of drug candidates and related information)
• Evaluation of our planned workflow and technology.
• Valuable feedback.
• Questions are welcome.
In an example project, the partner may provide the SMILES code of a compound and data characterizing such compound. Outcomes will be provided as follows: (i) predicted on-target and off-target protein binding, (ii) report on significantly perturbed protein complexes and signaling pathways in multiple human tissues, (iii) list predicted side effects in humans due to the above perturbations as MedDRA Preferred Terms. - Type and Size of Partner
- Big company
- SME 50 - 249
- SME 11-49
- R&D Institution
- University
- SME <=10
- Type of partnership
- Research and development cooperation agreement
Dissemination
- Technology keywords
- 06003001 - Bioinformatics
- 06001015 - Pharmaceutical Products / Drugs
- 01003006 - Computer Software
- 01003016 - Simulation
- Market keywords
- 02007012 - Medical/health software
- 05007002 - Pharmaceuticals/fine chemicals
- 02007016 - Artificial intelligence related software
- 04014 - Bioinformatics
- Sector Groups Involved
- Health
- Targeted countries
- All countries