Simulations Plus and Polish Academy Scientists Validate AI-Powered Drug Design Models in Breakthrough Discovery Targeting Inflammation and Immune Regulation
Simulations Plus, Inc. (Nasdaq: SLP), a prominent developer of modeling and simulation software for the biopharmaceutical industry, in collaboration with the Institute of Medical Biology of the Polish Academy of Sciences (IMB PAS), has announced a major milestone in their joint drug discovery initiative. The results of their artificial intelligence-driven drug design (AIDD) research, which leverages the ADMET Predictor® platform, have been published in the prestigious journal ACS Medicinal Chemistry Letters, a peer-reviewed publication by the American Chemical Society.
The publication represents a significant validation of Simulations Plus’s computational platform, particularly the AIDD module within ADMET Predictor®, as a robust and reliable tool for streamlining early-stage drug development. This successful collaboration demonstrates how AI-enhanced multi-parameter optimization can accelerate the design and validation of novel, bioactive drug candidates—ushering in a new era of rapid and informed therapeutic discovery.
An International Collaboration Focused on Targeting RORγ/RORγT
Launched in 2023, the partnership between Simulations Plus and IMB PAS was centered on discovering new modulators of the RORγ and RORγT receptors. These nuclear receptors play key roles in regulating gene expression related to inflammation and immune response, making them attractive targets for treating autoimmune disorders, inflammatory diseases, and certain cancers. The RORγT isoform, in particular, is a critical driver of Th17 cell differentiation and the secretion of pro-inflammatory cytokines such as IL-17, IL-21, and IL-22.
The core objective of the collaboration was to design novel small-molecule ligands capable of modulating RORγ/RORγT activity. The teams employed the AIDD module within ADMET Predictor®—a cutting-edge software suite developed by Simulations Plus that integrates machine learning, cheminformatics, and physiologically-based pharmacokinetic (PBPK) modeling—to generate, evaluate, and optimize potential lead compounds.
Remarkably, within a span of just three months, the two teams had designed, optimized, synthesized, and tested new compounds targeting RORγ/RORγT. Their approach combined predictive modeling of potency, absorption, metabolism, and toxicity profiles with synthetic accessibility and drug-like property assessments. The speed and success of this cycle reflect the power of AI-enabled drug design tools when strategically applied to real-world therapeutic challenges.
High Success Rate and Novel Chemistry Confirm Model Accuracy
According to Rafal A. Bachorz, PhD, Senior Principal Applied Scientist at Simulations Plus and lead author of the publication, the Simulations results were highly encouraging. Among the 27 compounds synthesized and tested, approximately 70% demonstrated meaningful inhibition of RORγT activity in vitro—closely matching or even surpassing the predicted activity scores from the ADMET Predictor® platform.
“Our lead compound exhibited strong inverse agonist activity with a novel indolizine scaffold that had not been previously reported in the context of RORγT inhibition,” said Dr. Bachorz. “This scaffold is of particular interest, not only due to its high potency but also because it opens a new chemical space for future exploration. The compound showed efficacy in cellular assays, lacked significant cytotoxicity, and significantly suppressed pro-inflammatory Th17 cytokine expression in human T cells.”
Further in vitro ADMET profiling revealed that the most potent compound also possessed favorable pharmacokinetic and drug-like properties. These included strong permeability, low predicted toxicity, and adequate metabolic stability—indicators that the molecule could progress into further rounds of preclinical optimization and potentially, in vivo testing. The congruence between experimental outcomes and predicted profiles reinforces the reliability of the AIDD module as a predictive tool in medicinal chemistry.
AI and Machine Learning as Catalysts for Efficient Drug Discovery
For Simulations Plus, the results reinforce their broader mission: to empower drug discovery and development teams with digital tools that enable faster, more precise, and cost-effective innovation. Their flagship product, ADMET Predictor®, is a state-of-the-art computational platform that incorporates artificial intelligence and machine learning algorithms to predict over 175 properties of chemical compounds—ranging from physicochemical traits to ADMET (Absorption, Distribution, Metabolism, Excretion, and Toxicity) profiles.
The AIDD module within ADMET Predictor® is specifically tailored to facilitate multi-parameter optimization during the lead discovery and lead optimization phases. This includes de novo molecular design, activity prediction, property balancing, and scaffold generation—all within an iterative framework that aligns computational insight with medicinal chemistry intuition.
According to Dr. Viera Lukacova, Chief Scientific Officer at Simulations Plus, these capabilities are particularly valuable when exploring novel targets like RORγT. “ADMET Predictor and its AIDD module offer our clients and partners a first-to-invent advantage by enabling rapid and intelligent compound design,” she said. “We’re especially pleased to see our platform validated in the context of a promising immuno-inflammatory target. The collaboration with IMB PAS highlights what’s possible when computational modeling is paired with top-tier academic research and medicinal chemistry expertise.”
Dr. Lukacova also emphasized the broader implications of this success. “Beyond this specific program, our approach has the potential to transform the early-stage discovery landscape, reducing reliance on trial-and-error synthesis and instead allowing teams to focus on the most viable, high-potential candidates. We look forward to continuing our collaboration with IMB PAS and further advancing research in immune-related disorders.”
Future Directions: Optimizing Leads and Exploring Therapeutic Applications
The promising results from this initial round of compound synthesis and testing have laid the foundation for continued exploration of RORγ/RORγT modulation. IMB PAS and Simulations Plus plan to extend their collaboration with additional rounds of scaffold optimization, structure-activity relationship (SAR) expansion, and refinement of pharmacokinetic profiles.
One of the most exciting aspects of this research is its translational potential. RORγT inhibitors are being investigated for a wide range of clinical indications, including multiple sclerosis, psoriasis, rheumatoid arthritis, and inflammatory bowel disease, as well as certain cancers where immune regulation plays a key role in tumor progression and evasion. The discovery of a new class of RORγT inverse agonists, such as the indolizine-based compound identified in this study, could provide a novel therapeutic avenue for these challenging diseases.
Moreover, the successful validation of AI-driven prediction models lends additional credibility to in silico approaches for drug design. As pharmaceutical and biotech companies face increasing pressure to improve efficiency, reduce development costs, and bring drugs to market faster, the use of predictive software tools like ADMET Predictor® is expected to grow significantly.
Bridging Academia and Industry Through AI Innovation
The collaboration between Simulations Plus and IMB PAS serves as a model for how academic institutions and industry partners can synergize their strengths to advance scientific discovery. By combining computational resources, software tools, and biological expertise, both organizations were able to accelerate the path from hypothesis to validated compounds in just a few months.
Such partnerships also highlight the democratization of drug discovery made possible by digital tools. Smaller research institutions can now leverage AI and machine learning to conduct high-quality drug development programs that were previously limited to major pharmaceutical players with vast resources.
The published study not only affirms the scientific value of the work but also serves as a benchmark for future collaborations that aim to harness AI for therapeutic innovation. It provides a replicable roadmap for other research teams aiming to blend computational design with experimental validation in a seamless and efficient workflow.
About Simulations Plus
Simulations Plus, Inc. is a leading provider of modeling and simulation software and consulting services to support drug discovery and development. Their flagship products include ADMET Predictor®, GastroPlus®, Monolix®, and DILIsym®. The company partners with pharma, biotech, and regulatory agencies worldwide to optimize therapeutic development using in silico approaches.
About the Institute of Medical Biology of the Polish Academy of Sciences (IMB PAS)
IMB PAS is a leading scientific institution in Poland dedicated to research in immunology, molecular biology, and medical biotechnology. The institute fosters cross-disciplinary collaborations and has contributed significantly to the understanding of inflammation, cancer biology, and immune-mediated diseases.
