Schrödinger APAC Webinar Series, Virtual
We are excited to be hosting the Schrödinger APAC Webinar Series – over 6 weeks we will host 6 talks by expert panelists from Schrödinger and universities. In this series, we will explore how physics-based molecular modeling and state-of-the-art ML/AI tools are impacting the drug discovery pipeline, from virtual screening in hit discovery through property predictions in pharmaceutical formulation. By registering for the series you will have access to watch all 6 webinars.
Date | Presentation Title Abstract | Speaker Chairperson |
September 6th | Discovering drug molecules using the Schrödinger Platform – Overview and case studies Abstract: Computational techniques are widely adopted in understanding the structure-function relationship and for screening millions of molecules for lead identification and optimisation. In recent years many techniques have emerged in virtual screening large datasets of molecules, predicting the binding affinity of these molecules, optimizing ADME properties, etc. The Schrödinger molecular design platform covers methods ranging from ultra-fast but approximate approaches to advanced physics-based in silico simulation methods to identify clinical candidates. This talk will highlight the recent development in computational methods used to identify a few hundred molecules from datasets of billions of molecules. The results across several successful drug discovery programs will be presented. | Dr. Pritesh Bhat, Principal Scientist, Schrödinger Chairperson: |
| September 13th 5:00pm SGT 6:00pm JST | Virtual screening approaches for small molecule drug discovery Abstract: Computer-aided drug discovery (CADD) plays a significant role in the discovery of new drugs in preclinical drug design. Modern computational techniques and algorithms have greatly expedited and enhanced the identification of potential drug candidates for emerging novel targets in the pharmaceutical industry. CADD involves two principal approaches: structure-based and ligand-based virtual screening techniques, which include molecular docking, pharmacophore modeling, de novo design, QSAR, and molecular dynamics. Researchers and scientists debate and discuss the rationality of computational tools despite the widespread use of these tools and the extensive literature that supports them. Therefore, determining when and how to utilize these molecular modeling techniques can be very interesting. This talk will discuss the fundamentals and applications of applying computational methods in drug discovery design and optimization processes to gain valuable information. This will allow researchers to maximize their efficiency by concentrating their experimental efforts on the most promising candidates. | Mr. Vinod Devaraji, Senior Scientist, Schrödinger Chairperson: |
| September 20th 5:00pm SGT 6:00pm JST | Optimizing protein therapeutics using accurate physics-based approaches Abstract: Therapeutic proteins are a crucial class of medicines that can treat diseases that are difficult to target with small molecules. However, the properties of the proteins often need to be optimized including binding affinity and thermal stability. In this talk, we will discuss the application of free energy perturbation in protein engineering. We will show that free energy perturbation can accurately predict thermal stability and binding affinity. In combination with the faster MM-GBSA based method, the screening workflow can quickly identify most promising mutations with only a fraction of experimental effort. In addition, we will showcase the possibility of predicting pH-dependent binding. | Dr. Jianxin Duan, Fellow, Schrödinger Chairperson: |
| September 27th 5:00pm SGT 6:00pm JST | Practical implementation of the hit-to-lead process with a focus on FEP+: A case study of AKT2 inhibitor development Abstract: In recent years, rapid advancements in computational science have made it possible to utilize advanced computational methods in the drug discovery process. FEP+ is one such computational method that allows for the prediction of protein-ligand binding free energy with far greater accuracy than conventional ligand evaluation methods like docking scores or MM-GBSA. Schrödinger has actively shared information about the latest achievements derived from our in-house drug discovery process, centered around FEP+. However, due to ongoing development projects, we haven't been able to provide detailed insights into the specific structural transformations from hit compounds to lead candidates. Therefore, in this presentation, we will use the development of an AKT2 inhibitor reported in "J. Med. Chem., 2007, 50, 2293-2296" as an example. We will showcase the process from acquiring fragment hits against the target to lead compound generation, reconstructed using FEP+ as the primary design tool. Throughout the presentation, we will introduce the usage of FEP+ in various steps, such as evaluating multiple hit compounds as chemical starting points, obtaining initial structure-activity relationship (SAR) information, and exploring synthesis approaches to achieve lead compounds. Additionally, we will discuss the specific application of advanced computational tools like WaterMap and QM pKa calculations to deal with ADMET issues. | Dr. Osamu Ichihara, Senior Principal Scientist, Schrödinger Chairperson: |
| October 4th 5:00pm SGT 6:00pm JST | Leveraging machine learning applications combined with physics-based modeling for drug discovery Abstract: Machine learning strategies in drug discovery are becoming increasingly popular and can be used in various areas. In the Schrödinger Suite DeepAutoQSAR serves as the main tool for training machine learning models to predict activity, ADMET, and other compound properties. In order to leverage both the proven accuracy and wide applicability domain of physics-based computational models, such as QM and FEP, together with the speed and scale of machine learning, we have combined our physics-based modeling technologies with an active learning framework. This framework can effectively speed up virtual screening methods such as in Active Learning -Glide, Active Learning-FEP, and Active Learning-ABFEP, or to improve the accuracy and applicability domain of models such as pKa prediction in Epik and machine learned force fields such as QRNN. We will also discuss how to utilize machine learning protein structure prediction methods to enable new targets for structure-based drug design. | Dr. Marton Vass, Principal Scientist, Schrödinger Chairperson: |
| October 11th 5:00pm SGT 6:00pm JST | Solving pharmaceutical formulation problems using molecular modeling Abstract: Computer-aided formulation design is an emerging field of great interest, attracting researchers worldwide. The pharmaceutical industry displays a keen enthusiasm for unraveling the underlying molecular mechanisms that could enable them to predict the physical and chemical stability of pharmaceutical formulations. In this context, Schrödinger's Materials Science Suite (MS Suite) emerges as a distinctive toolkit with the capacity to tackle various formulation-related challenges. MS Suite offers automated workflows designed to comprehend a multitude of physicochemical properties associated with pharmaceutical formulations. It achieves this by harnessing advanced physics-based simulation techniques such as quantum mechanics, molecular dynamics, and coarse-graining. These approaches provide invaluable insights from a molecular perspective, offering a deeper understanding of the complexities involved. During this presentation, we will delve into successful case studies that address challenges related to solubility and stability in pharmaceutical formulations. These case studies will exemplify how MS Suite has been instrumental in identifying optimal formulation conditions, mitigating degradation pathways, and enhancing overall product quality. | Dr. Sudharsan Pandiyan, Principal Scientist, Schrödinger Chairperson: |