Scientific Abstract

G protein–coupled receptors (GPCRs) are among the most extensively studied protein families in pharmacology and drug discovery, yet many receptors remain poorly characterized or “orphan,” lacking known endogenous ligands. In this conversation, Charlotte Crauwels discusses her work developing computational strategies to support the design of hybrid or chimeric GPCRs.

Charlotte Crauwels’ research focuses on building computational pipelines that guide the rational design of GPCR chimeras by combining structural elements from well-characterized receptors with poorly understood ones. These hybrid receptors enable researchers to probe receptor signaling, investigate G protein coupling preferences, and explore structural determinants of receptor activation. By evaluating large numbers of candidate constructs in silico, computational screening can help prioritize experimental designs before testing in the laboratory.

The discussion also explores the broader relationship between experimental and computational research in GPCR biology. Ms. Crauwels emphasizes the critical role of well-annotated experimental data for training predictive models and highlights the importance of collaboration between bench scientists and computational researchers. Listeners gain insight into how computational protein design can accelerate receptor signaling research and contribute to new approaches for GPCR discovery and pharmacological characterization.

About the Guest

Charlotte Crauwels is a computational scientist specializing in protein design and GPCR biology. She conducted her PhD at the Vrije Universiteit Brussel in the research group of Dr. Wim Vranken, focusing on computational approaches for GPCR engineering. Her research centers on developing computational pipelines that guide the design of hybrid or chimeric GPCR constructs to study receptor signaling and structural determinants of receptor activation. Her work integrates bioinformatics, protein modeling, and experimental collaboration to support GPCR discovery and functional characterization.

Scientific Themes of the Conversation

• Computational protein design for GPCR engineering

• Hybrid and chimeric GPCR constructs

• Experimental–computational feedback loops in receptor biology

• Data quality and annotation in computational pharmacology

• Deorphanization strategies for poorly characterized GPCRs

• Structural determinants of GPCR signaling pathways

Key Insights from the Conversation

Experimental Data Drives Computational Discovery

Computational models are fundamentally dependent on high-quality experimental data. Charlotte Crauwels emphasizes that predictive models are only as reliable as the datasets used to train them, highlighting the importance of properly annotated experimental datasets in GPCR pharmacology.

Computational Tools Help Guide Experimental Design

Rather than replacing experiments, computational approaches provide guidance. In Ms. Crauwels’ work, computational pipelines rank potential GPCR chimera designs, helping experimental scientists prioritize which constructs to test first in the lab.

Hybrid GPCRs Provide Insight into Orphan Receptors

Chimeric GPCR constructs combine components from well-characterized receptors with poorly understood receptors. This strategy can reveal signaling pathways or structural features of orphan GPCRs that cannot easily be studied otherwise.

Collaboration Between Bench and Computational Scientists Is Essential

Computational researchers depend on experimental validation, while experimental scientists benefit from predictive models that help narrow down possible hypotheses. Successful GPCR research increasingly depends on close collaboration between these two domains.

Communication Challenges Between Research Disciplines

Computational and experimental scientists often use different terminology and assumptions. Dr. Crauwels describes how misunderstandings can arise from differences in language, experimental details, or sequence definitions—highlighting the need for clear communication in interdisciplinary projects.

Data Transparency Enables Better Models

Publishing negative results and detailed experimental metadata can significantly improve computational models. Even unsuccessful experiments can contain valuable information for machine learning or predictive modeling.

GPCR Biology Continues to Produce “Gotcha Moments”

Despite decades of research, GPCRs remain complex and surprising. Dr. Crauwels notes that researchers often experience unexpected results that challenge assumptions about receptor signaling and structural behavior.

Episode Timeline

00:00 — Introduction

Introduction of Dr. Crauwels and her PhD work in computational GPCR research.

02:20 — Scientific Path Into Computational Biology

How bioengineering and bioinformatics led Dr. Crauwels toward computational protein design.

09:15 — The Role of Computational Science in GPCR Research

The relationship between experimental data and computational modeling.

16:00 — Collaboration Between Experimental and Computational Scientists

Why feedback loops between lab work and computational analysis are critical.

26:30 — Designing Hybrid GPCR Constructs

The concept of chimeric GPCRs and how they can be used to study receptor signaling.

30:00 — Building a Computational Pipeline for Chimera Design

Using in silico modeling to rank candidate receptor designs before laboratory testing.

36:30 — GPCR ChimeraDB and Data Challenges

The challenge of collecting standardized data on chimeric GPCR experiments.

41:30 — “Gotcha Moments” in GPCR Research

Why GPCRs continue to surprise even experienced researchers.

43:50 — Advice for Young Computational Scientists

The importance of collaboration, conferences, and stepping outside research silos.

48:20 — Career Reflections and What Comes Next

Thoughts on the transition after a PhD and continuing research in GPCR biology.

Selected Quotes

Full Transcript

(Formatted for readability — original transcript preserved)

Yamina Berchiche:

Hello everyone. This is Yamina from Dr. GPCR, and today I'm very excited to have with me Charlotte Crauwels. Charlotte is finishing up her PhD, and we’ve been trying to sit down and have this conversation for quite some time. Charlotte, welcome to the podcast.

Charlotte Crauwels:

Thank you for having me.

Yamina Berchiche:

Why don't we start by you introducing yourself and telling us about what you're working on right now?

Charlotte Crauwels:

So as you mentioned, I’m a final-year PhD student at the Vrije Universiteit Brussel in Belgium. I work in the group of Dr. Wim Vranken. I would define myself as a computational protein designer focusing on GPCRs.

Looking back, my path to this work was not very linear. I wasn’t originally planning to do a PhD, and I wasn’t planning to work on GPCRs either. I studied bioengineering in Brussels, and during my master’s I was introduced to bioinformatics. I discovered that I really enjoyed working with data and computational tools.

For my master’s thesis, I studied antibodies and nanobodies and how they bind to their antigens. That experience convinced me that research might be something I wanted to pursue.

At that time, I wasn’t sure what career direction I wanted, so I started contacting computational research groups in Belgium. Eventually, I connected with my current supervisor and began developing GPCR-related research projects.

And by coincidence, I ended up working on GPCRs—and I’m actually very happy that I did.

Yamina Berchiche:

Looking back at your PhD trajectory, what originally drew you to computational research?

Charlotte Crauwels:

I think it was the idea that we now have enormous amounts of biological data available. We have powerful computational tools that allow us to explore biological systems in ways that were not possible before.

So my motivation was really the idea of engineering life—using data and computational methods to understand biological systems and potentially optimize them.

Yamina Berchiche:

Computational scientists are not always well represented in GPCR discussions. How do you see the role of computational approaches in GPCR biology?

Charlotte Crauwels:

I think computational scientists are very dependent on experimental work. We don't generate the original data ourselves—we analyze and build models based on experimental data.

That means the quality of computational work depends heavily on the quality of the underlying experimental datasets.

People often say “garbage in, garbage out.” If the experimental data is poorly annotated or incomplete, the computational predictions will not be reliable.

This is why sharing experimental data—and annotating it well—is so important.

Yamina Berchiche:

You also mentioned the importance of collaboration between experimental and computational researchers.

Charlotte Crauwels:

Yes, absolutely. I always encourage computational scientists to find experimental collaborators.

When you collaborate across disciplines, you create a feedback loop. Computational models generate predictions, experimentalists test them, and the new data helps refine the models.

It also makes research more dynamic and motivating. You feel that your work contributes directly to someone else’s experiments.

Yamina Berchiche:

Let’s talk about your PhD project. What was the core idea behind your research?

Charlotte Crauwels:

The goal of my PhD was to develop a computational pipeline that supports the design of hybrid—or chimeric—GPCRs.

A hybrid GPCR is a receptor constructed by combining elements from two different receptors. Typically, you take a well-characterized GPCR and combine it with a poorly characterized receptor.

For example, you might take the extracellular and transmembrane domains of the beta-2 adrenergic receptor and combine them with intracellular loops from an orphan GPCR.

This allows researchers to activate the receptor using a known ligand while observing the signaling behavior of the poorly characterized receptor.

Yamina Berchiche:

So you can essentially probe the signaling pathways of receptors that are otherwise difficult to study.

Charlotte Crauwels:

Exactly.

These hybrid constructs allow researchers to investigate signaling pathways, structural determinants of receptor activation, and even help resolve receptor structures.

But designing these hybrids is difficult. Some receptor combinations work very well, while others fail—even when the receptors are very similar.

My work focuses on building a computational framework that generates many possible hybrid designs and evaluates them in silico.

The system ranks candidate constructs and suggests the most promising ones for experimental testing.

Yamina Berchiche:

That seems much more efficient than testing dozens of constructs experimentally.

Charlotte Crauwels:

Yes, that’s the idea. My goal is simply to provide a tool that helps experimental scientists prioritize which constructs to test first.

Yamina Berchiche:

You also built a database during your PhD.

Charlotte Crauwels:

Yes. One of the challenges I faced early in my PhD was collecting data on previously published chimeric GPCR constructs.

The terminology used in the literature is inconsistent. Some papers call them hybrid receptors, others call them chimeric receptors, and sometimes they’re just described as constructs.

It took over a year to gather and curate the data.

We eventually created a publicly available database called GPCR ChimeraDB, which collects known hybrid GPCR designs and links them to structural and sequence information.

Yamina Berchiche:

Let’s switch to a more personal question. What advice would you give to scientists entering computational GPCR research?

Charlotte Crauwels:

First, dare to ask.

People are often more willing to help than we think. Ask to attend conferences, ask for collaborations, ask questions.

Second, don’t stay in your bubble.

Computational scientists can easily become isolated. It’s important to attend workshops, meet experimental scientists, and keep the broader research context in mind.

Yamina Berchiche:

Final question. What comes next after your PhD?

Charlotte Crauwels:

I’m not completely sure yet. But I would be very happy if I can continue working on GPCRs in some way.

I feel like I’m just starting to understand how research works. After several years of learning and making mistakes, you finally feel ready to take on bigger questions.

So whether it’s in academia or industry, I hope to continue doing research related to GPCR biology.

Yamina Berchiche:

Charlotte, thank you so much for joining the podcast. It was a pleasure speaking with you.

Charlotte Crauwels:

Thank you. It was very fun.