AAPS PharmSci360
Date & Time - November 9th-12th, 2025
Location - San Antonio, Texas
Schrödinger is excited to be participating in the AAPS PharmSci360 conference taking place on November 9th – 12th in San Antonio, Texas. Stop by booth #3329 to speak with Schrödinger scientists. Join us for presentations by Schrödinger scientists.
NOV 12 | 9:30AM 
Stars at Night B1 Selecting Polymer Excipients Using Molecular Dynamics Simulation
Speaker:
Ben Coscia, Principal Scientist II, Schrödinger
Abstract:
The presentation will begin with a description of polymer excipients and formulation systems with a focus on the varying molecular architectures that are present in the commercially available pharmaceutical formulation polymers. This will be followed by details of the balance of mechanisms necessary for a successful selection of a polymer for amorphous solid dispersion or encapsulation vehicle to maintain stability on the shelf while allowing for optimum release of the drug.
The major highlight of the presentation will then be provided with a summary of modern computer simulation technologies with a focus on bottom up techniques to enable down selection from a set of major commercial polymer excipients to provide the best combination of properties for a given active pharmaceutical ingredient. Examples will be provided for this selection process for specific drug – excipient combinations.
The key factors that are included in the screening tools will be reviewed with additional focus on:
• Important considerations not included in the analysis
• Areas of applicability
• Accuracy considerations and appropriate stage in development for inclusion of method, including regulatory concerns
Learning Objectives:
• Explain how molecular interaction connects to polymeric formulation considerations.
• Identify screening applications where simulation could provide initial downselection.
• Understand the expected time-scales and required input data for successful integration of simulation into formulation development.
NOV 12 | 9:30AM
Stars at Night B2/B3 Computational Modeling for Lipid Nanoparticle Formulation Development, Efficacy and Targeting
Speaker:
John Shelley, Fellow, Schrödinger
Abstract:
We will start off describing lipid nanoparticle – nucleic acid (LNPna) drug formulation, technology and delivery
strategies along with a problem statement highlighting – how the limited understanding of the structure and
behavior of LNPna systems during the formulation development, storage, and in vivo delivery negatively
impacts the overall viability of a drug program. The understanding of the structural aspects of the LNPna
helps in the selection of the excipients, encapsulation efficiency, efficacy, improvement in the drug targeting
and hence the development for new therapeutic areas.
Next we will describe a number of computer modeling technologies, particularly physics-based approaches,
for characterizing lipid nanoparticle behavior and structure at the intraparticle – molecular interaction, whole
particle and particle-environment levels to enable more effective formulation, drug production and delivery
strategies. Illustrative examples of the simulation of the self-assembly of LNPna structures for actual
formulations will be part of the talk.
Distinct case studies, illustrating how drug products would be influenced when: 1. The compositional and pH dependence of the LNP structure is unknown, 2. Seeking to improve the endosomal release of the mRNA, 3. Calculating the apparent pKa values of ionizable lipids, 4. There are challenges with incorporating longer nucleic acid sequences into the LNP.
Additionally, using modeling to inform targeting research will be discussed. The studies will illustrate the
complementary use of a number of computational technologies to address these LNPna challenges, including all-atom simulations, coarse-grained simulations, molecular mechanics, ML and quantum
Submission Title: Computational modeling for lipid nanoparticle formulation development, efficacy and targeting
mechanical.
The methodology and current limitations thereof will also be described with examples, namely, 1. Relevant issues that are not addressed, 2. Applicability (complexity) and methodology limitations, 3.Accuracy and reliability
Learning Objectives:
• Understand the main challenges in LNPna research and development and how gaps in our knowledge limit progress in improving efficacy and diversification of ailments treated and treatment strategies.
• Identify projects for which computation tools in combination with experimental techniques can contribute to formulation selection and the optimization of the manufacturing process.
• Know the limitations of computational techniques in order to assess when to apply them, their potential impact, and how to interpret the results.
NOV 12 | 1:30PM
Start at Night B1 Keynote: Accelerating Drug Discovery and Development with Advanced Computational Modeling
Speaker:
Robert Abel, Executive Vice President, Chief Science Officer, Platform, Schrödinger
Abstract:
The efficient discovery and development of high-quality clinical candidates remain hampered by late-stage failures, often arising from unforeseen toxicity or suboptimal physicochemical properties. To address these persistent challenges, we are developing and deploying an integrated suite of advanced computational tools designed to accelerate the design-make-test-analyze cycle and increase the probability of success. This presentation will detail several new and emerging scientific advances that are transforming the path from initial hit to a viable development candidate. Central to our strategy is the development of predictive, structure-based in silico ADMET panels, conceived as computational analogues to standard experimental assays. These models allow for the critical assessment of off-target liabilities and other key properties early in the discovery process. We then leverage these predictive models within an ultra-large-scale de novo design platform to generate and prioritize novel molecular structures optimized against multiple objectives simultaneously. A case study on Wee1 kinase will be presented to highlight how this integrated approach was used to successfully ideate novel matter and resolve a challenging kinome-wide selectivity issue. Finally, we demonstrate the extension of these technologies beyond candidate selection to address critical drug development challenges. We will present cutting-edge methods for prospectively predicting small molecule crystal structure polymorphs and their relative stabilities, along with techniques for accurately estimating polymorph-specific aqueous solubility. Collectively, these integrated computational strategies provide a powerful, holistic approach for accelerating the design, optimization, and successful development of promising new medicines.
Learning Objectives:
• Recognize how predictive, structure-based in silico panels can be used to assess ADMET and toxicity risks early in drug discovery.
• Understand the application of ultra-large-scale de novo design to overcome specific project challenges, such as kinase selectivity.
• Appreciate how computational methods can be extended to address late-stage development hurdles like crystal structure polymorphism and solubility prediction.