PolyModels Hub

🚀 Big news, friends. BIG NEWS. 🎉 We're proud to announce our successful seed funding round of GBP 1.1M by Marathon Venture Capital. We couldn't have done it without our earliest adopters, followers, supporters, and friends. At PolyModels Hub, we are revolutionizing pharma process development by empowering scientists with advanced digital tools and insights to enhance every step of drug development and manufacturing. If you're part of our community - please consider us as your partner in 2024 to dramatically reduce development time and optimize results for your drug development teams. Check out the article and the full press release for more details: https://lnkd.in/dPgpRj8d Cheers!

🔍 Building models is just the beginning when it comes to leveraging them effectively in the pharmaceutical industry. From the very first step, it's crucial to align your modeling efforts with regulatory considerations, even when regulatory inclusion and approvals may seem far off. 🚀 At PolyModels Hub, we’re developing workflows and software solutions that help you seamlessly meet technical, regulatory, and business needs—ensuring your focus remains on delivering value. 📅 Join us for an exclusive webinar in partnership with VIAVI MicroNIR, where we’ll dive deep into a practical PAT example, exploring how to navigate the complexities of process analytics. Don’t miss this opportunity to gain insights into bridging technical innovation with regulatory strategy! 👉 Register now and stay ahead in the industry!

🔍💊 Developing a High-Fidelity DEM Digital Twin for Continuous Direct Compression 🛠️💻 This work presents the development of a high-fidelity digital twin using the Discrete Element Method (DEM) to simulate continuous direct compression processes in drug product manufacturing. By employing typical pharmaceutical equipment and materials, the platform supports the design and testing of control strategies. The study from Dalibor Jajcevic, Johan Remmelgas, Peter Toson, Marko Matić, Theresa R. Hörmann-Kincses, Michela Beretta, Jakob Rehrl, Johannes Poms, Thomas O'Connor, Abdollah Koolivand, Geng (Michael) Tian, Scott Krull, and Johannes Khinast focuses on material characterization, DEM model development, and a rigorous calibration workflow that ensures accurate predictions across various processes. 🧩 Key Insights: 🔬 Discrete Element Method (DEM) Calibration: A cohesive DEM contact model calibration was introduced, based on bulk density, compression, shear cell, and rotating drum tests. ⚙ High-Fidelity Simulations: The calibration workflow demonstrated accurate predictions of material behavior under varying stress states. 🔄 Process Control: The digital twin effectively differentiated between material batches, showcasing its potential for improving process control in continuous direct compression. 📊 Predictive Accuracy: Small-scale material tests proved valuable for predicting residence time distribution in continuous manufacturing. 📚 Link to Publication: https://lnkd.in/dUiVAPJD #DigitalTwin #ContinuousManufacturing #DEM #ProcessOptimization #PharmaceuticalEngineering #DirectCompression #PolyModelsHub

🔍💊 MLAPI: Machine Learning Framework for Automated Drug Particle Synthesis in Continuous Flow 🤖🔄 Machine learning (ML) models are gaining traction in pharmaceutical manufacturing, particularly within process analytical technology (PAT) frameworks. However, their integration often faces challenges due to extensive data requirements. This work by Arun Pankajakshan, Sayan Pal, Nicholas Snead, Juan Almeida, Maximilian Besenhard, Shorooq Abukhamees, Duncan Craig, Asterios Gavriilidis, Luca Mazzei, and Federico Galvanin presents MLAPI, a data-efficient machine learning framework to guide drug particle synthesis in automated continuous flow platforms. 🔬 The framework employs classification algorithms to pinpoint fouling-free operating regions, coupled with a multiple-output Gaussian process (GP) regression model that connects key process parameters to drug particle size. Active learning is integrated to strategically generate new data for the GP model, enabling precise control over particle synthesis with minimal data input. 🧩 Key Insights: 🤖 Data-Efficient Machine Learning: MLAPI reduces big data dependency through the application of active learning, optimizing data generation for GP model training. 💡 Process Control via ML: Enables fine-tuning of drug particle size by relating key process parameters to synthesis outcomes, ensuring targeted bioavailability and processability. 🔄 Automated Continuous Flow: Demonstrates the successful integration of ML models into an automated flow precipitation platform, improving operational efficiency and fouling-free regions. 🧪 Practical Application: Validated through the synthesis of ibuprofen microparticles, highlighting the real-world utility of the framework. 📚 Link to Publication: https://lnkd.in/d4GkgDZU #MachineLearning #PharmaceuticalManufacturing #PATFramework #ContinuousFlow #DrugParticleSynthesis #GaussianProcess #Ibuprofen #Bioavailability

🔍🧬 Machine Learning for Productive CHO Cell Line Selection in Biopharmaceutical Development 🤖💉 The identification of highly productive CHO cell lines is vital for optimizing monoclonal antibody (mAb) production in biopharmaceutical processes. This study by Gianmarco Barberi, (our own) Antonio Benedetti, Paloma Diaz Fernandez, Daniel C. Sevin, Johanna Korbeń, Gary Finka, Fabrizio Bezzo and Pierantonio Facco introduces a novel machine learning approach that leverages dynamic metabolomic data from the Ambr®15 scale to support early-stage cell line selection. 🔬 By analyzing metabolomic profiles, the procedure identifies key biomarkers and metabolic pathways that correlate with high mAb productivity, enabling more informed decisions during bioprocess development and scale-up. Metabolomic dynamics reveal insights into cell line performance, with early stages dominated by tricarboxylic acid cycle pathways and late stages influenced by amino and nucleotide sugar pathways. This approach allows for early identification of productive cell lines and supports the discovery of crucial metabolic pathways linked to mAb production. 🧩 Key Insights: ⚙ Machine Learning Integration: A robust ML procedure analyzes metabolomic data to identify high-productivity CHO cell lines early in the development process. 🧬 Biomarker Discovery: Key biomarkers and metabolic pathways influencing mAb productivity are revealed, providing actionable insights for process development. ⏱️ Early Selection: Enables early identification of productive cell lines, streamlining the bioprocess development and scale-up. 🔄 Metabolic Pathway Insights: Highlights the influence of tricarboxylic acid cycle pathways in early cultivation stages and amino/nucleotide sugar pathways in later stages. 📚 Link to Publication: https://lnkd.in/dM7K3EaN #Bioprocessing #MachineLearning #CHOCells #MonoclonalAntibodies #Metabolomics #CellLineSelection #Biopharmaceuticals #Productivity

At PolyModels Hub, we’re proud to work alongside such innovative partners, helping to push the boundaries of technology in pharma development and manufacturing. It’s moments like these that remind us of the importance of collaboration in delivering the best solutions for patients worldwide!

🔍🔬 Comparative Assessment of Simulation-Based vs. Surrogate-Based Flowsheet Optimization 🔄🧪 In process design and optimization, achieving a balance between computational efficiency and solution accuracy is crucial. This study by Niki Triantafyllou, Ben Lyons, Andrea Bernardi, Benoit Chachuat, Cleo Kontoravdi and Maria Papathanasiou proposes a framework for reduced-space Bayesian optimization of process flowsheets, comparing simulation-based and surrogate-based approaches. Through global sensitivity analysis, the framework reduces dimensionality by identifying critical process variables that significantly impact key objectives like productivity and operating costs. 🔬 The study evaluates these optimization techniques on two case studies: plasmid DNA production in a biopharmaceutical simulator and dimethyl ether (DME) production in a chemical process simulator. The comparison reveals that while simulation-based Bayesian optimization yields more accurate objective function values, surrogate-based methods offer greater computational efficiency. 🧩 Key Insights: 🔄 Dimensionality Reduction: Uses global sensitivity analysis to identify critical variables, simplifying the optimization space. ⚙️ Simulation vs. Surrogate: Assesses both approaches, showing simulation-based optimization achieves better performance outcomes, while surrogate-based optimization excels in computational efficiency. 🧪 Biopharma and Chemical Processes: Applied to the production of plasmid DNA and dimethyl ether, demonstrating versatility across industries. 📊 Optimization Trade-Offs: Highlights the trade-offs between computational effectiveness and solution accuracy, providing insights into the optimal strategy for different scenarios. 📚 Link to Publication: https://lnkd.in/dVKuuVAv #ProcessOptimization #BayesianOptimization #SimulationModels #SurrogateModels #DimensionalityReduction #Biopharmaceuticals #ChemicalProcesses #DigitalTools

🔍🔬 Introducing GlyCompute: Automated Analysis of Protein N-Linked Glycosylation Kinetics 📊💻 Understanding glycosylation pathways is essential for advancing the field of glycosciences, yet the in silico analysis of glycosylation kinetics remains underexplored. In response to this gap, Konstantinos Flevaris, Pavlos Kotidis & Cleo Kontoravdi developed GlyCompute , an open-source framework designed to streamline the analysis, simulation, and parameterization of kinetic models for N-linked glycosylation, providing a more accessible, automated solution for glycoscientists. 🔬 This study introduces GlyCompute’s ability to generate and simulate glycosylation reaction networks based on experimentally observed N-glycan structures. The framework employs a Bayesian inference-based sequential parameter estimation strategy, automatically adjusting to match observed glycoprofiles, enabling efficient and accurate modeling. 🧩 Key Insights: 🔄 Automated Network Assembly: GlyCompute generates kinetic models by assembling reaction networks based on observed N-glycan structures and their abundances. 🤖 Bayesian Inference for Parameter Estimation: Utilizes a sequential approach to parameter estimation, ensuring the model fits the experimental data. 🔬 Application to CHO Cell Culture: Demonstrates strong agreement between the model and experimental data in a case study on protein N-linked glycoprofiles from CHO cell culture. 🛠️ Open-Source Flexibility: Available on GitHub, providing glycoscientists with a valuable tool for glycosylation kinetics analysis and enabling further development and collaboration. 📚 Link to Publication: https://lnkd.in/dVDC33ge 📂 Explore GlyCompute on GitHub: https://lnkd.in/dz9_3JwS #Glycoscience #GlycosylationKinetics #OpenSource #BayesianInference #ProteinAnalysis #Biomanufacturing #DigitalTools

We are #hiring!  Do you know someone who would be a great fit?

🔍🔥 Hot-Melt Extrusion: Advancements in Pharmaceutical Formulation and Continuous Manufacturing 💊🔄 Hot-melt extrusion (HME) has emerged as a globally recognized technology, transforming pharmaceutical manufacturing by enhancing the bioavailability of poorly soluble APIs and enabling continuous production. At the heart of this innovation is the twin-screw extruder (TSE), a versatile and customizable tool capable of handling various compounding and granulation processes. 📈 This review by Hemlata Patil, Dr. Sateesh Kumar Vemula, Sagar Narala, Preethi Lakkala, Siva Ram Munnangi, Nagarjuna Narala, Miguel O. Jara, Robert Williams III, Hibreniguss (Hibre) Terefe, Ph.D. and Michael Repka goes into the latest challenges, models, and strategies for scaling up HME, while exploring its growing applications in cutting-edge pharmaceutical technologies such as dry powder inhalers, 3D printing, PAT and amorphous solid dispersions. 🧩 Key Insights: • ⚙️ Twin-Screw Extruder Versatility: Outlines the adaptability of TSE for applications beyond traditional dosage forms, including nanoextrusion and 3D printing. • 🔄 Continuous Manufacturing: Emphasizes the advantages of continuous production in improving efficiency and scalability for pharmaceutical formulations. • 📊 PAT and Regulatory Considerations: Explores the role of PAT in ensuring process quality and compliance with regulatory standards during HME scale-up. • 🧬 Innovation Through Collaboration: Highlights the critical role of industry and academia in driving advancements in HME technology over the past decades. • 🚀 Future Applications: Investigates new and emerging uses for HME in novel formulations, broadening its potential for pharmaceutical development. 📚 Link to Publication: https://lnkd.in/d5xQVwPw #PharmaceuticalManufacturing #HotMeltExtrusion #ContinuousManufacturing #TwinScrewExtruder #ProcessOptimization #PAT #Bioavailability #PharmaInnovation #PolyModelsHub