An image of a person fishing represents virtual screening.

A Hands-On Introduction to High Throughput Virtual Screening

The Schrödinger Online Learning Program has launched a new course called High-Throughput Virtual Screening for Hit Finding and Evaluation that teaches how industry-leading computational molecular modeling tools are used in High Throughput Virtual Screening (HTVS) to accelerate hit finding and evaluation. We spoke with Schrödinger’s Education Specialist, Dr. Abigail Emtage and Director of Education, Dr. Jennifer Chambers, to learn more about this course.

Can you briefly explain some of the core concepts a participant will learn in this course?

First and foremost, we wanted to make sure that participants in this course would walk away feeling empowered to run their own HTVS calculations on their own structures. With this in mind, we start the course highlighting the relevance of virtual screening, the differences between various kinds of virtual screening techniques, and the benefits of screening very large ligand libraries. Then, we review how to evaluate, test, and validate their starting structures — both the protein receptors and ligands. After this, participants will run a virtual screen on a 100K ligand library and then learn how to analyze the results, such as diversity analysis and clustering, and what computational techniques could come next. Finally, participants will put all these skills to use by performing a virtual screening campaign on a 2M ligand library and submitting two compounds for evaluation by Absolute Binding Free Energy Perturbation.

Is this course a good fit for participants who are new to molecular modeling as well as those who know computational methods? 

The HTVS course requires a working knowledge of Maestro — the platform used in the course. That said, participants don’t need to be an expert in computational chemistry. In fact, this is an ideal course for a medicinal chemist who is familiar with Maestro and wants to advance their design skills. Course participants should also be comfortable with biochemical and drug design terminology, willing to dive into some light command line work, and be able to dedicate regular time to the course throughout the session. As several of the calculations that will be run in the course take a few hours to even a day to complete, this is not a course that participants can wait until the last minute to complete. That being said, we put this course together with a working scientist in mind. Through internal and beta testing, we have made sure to pace the course appropriately and allow sufficient time for a full-time worker or student to complete the content.

If someone who is new to Maestro wants to take this course, they can work alongside the videos in the Maestro Training Portal to get up to speed. 

“Our goal is for participants to feel comfortable doing high-throughput virtual screening on their own systems, not just the systems used in this course.”

Can you describe how participants will use FEP+ in this course? 

The final module of the course gives participants the opportunity to work through a virtual screening case study and select hit compounds from ligand-based and structure-based screening results to take forward for absolute binding affinity prediction using FEP+. Course participants will generate force field parameters for their chosen compounds using Force Field Builder and then run the absolute binding FEP+ job. The final assignment involves uploading the output of the FEP+ job along with a brief discussion of the results and justification of hit choice.

What new skills can a participant expect to gain regarding HTVS once they finish the course?

Participants who complete the material in the HTVS course should expect to gain practical experience with setting up, validating, running, and analyzing the output of various calculations related to target validation, structure analysis, and virtual screening. They will also learn valuable post-processing techniques for digging into the results from virtual screens. Importantly, this course will also demonstrate several scenarios and highlight good questions to ask when applying to their own research. Our goal is for participants to feel comfortable doing high-throughput virtual screening on their own systems, not just the systems used in this course.

What’s next for Schrödinger Online Learning?

We have just launched Introduction to Computational Antibody Engineering which teaches how industry-leading computational molecular modeling solutions are used to aid in antibody discovery and design and how to incorporate these computational approaches into research projects. This course will also provide a chance to work hands-on with Schrödinger’s BioLuminate software. 

These courses fulfill a need that we’ve seen in both the academic and commercial fields. We’re excited to help scientists advance their knowledge in molecular modeling in a way that is flexible and fits a participant’s own schedule. 


Learn more about Schrödinger Online Learning here or get in touch via 

Schrödinger Editorial Team

Sign Up Today


Sign up to receive quarterly updates with the latest from Extrapolations.