GPCR Collaboration: From Models to Medicine

Dr. GPCR Podcast
Nov 3
5 min read

Watch Episode 175

When Jens Carlsson was a PhD student, he thought collaborations slowed science down. While experimentalists struggled for months with crystallography, he could pull a clean protein structure from the Protein Data Bank in a single afternoon and move on. Why wait for someone else’s bottleneck?

That perspective changed the moment he entered GPCR research. Suddenly, computational shortcuts weren’t enough. GPCRs demanded something different: the integration of modeling, medicinal chemistry, and pharmacology. Docking scores were meaningless without assays. Predictions went nowhere without synthetic expertise. And most importantly, no single lab could cover the full terrain alone.

Today, Carlsson is Professor of Computational Biochemistry at Uppsala University and one of the strongest advocates for an approach that too many labs still resist. In his view, collaboration is not a bonus or an optional supplement—it is the only way GPCR drug discovery works.

From Lone Modeler to Collaborative Architect

Carlsson’s lab is built to plug directly into a larger ecosystem. At its core is a team of about ten computational chemists working on molecular docking, molecular dynamics, virtual screening, and more recently, machine learning. Supporting them is a medicinal chemist who can design and synthesize new scaffolds when commercial libraries run out of options. But Carlsson doesn’t try to do everything under one roof. For receptor assays and the biological interpretation of ligands, he relies on a network of collaborators worldwide, each bringing specialized expertise in different GPCR targets.

The power of this design lies not in its independence but in its interdependence. Predictions from modeling inform chemistry. Chemistry fuels assays. Assay data flows back into models. The cycle only works because every part is connected.

Closing the Trust Gap

If collaboration is so effective, why do so many labs avoid it? Carlsson points to a deeper problem: credibility. Experimentalists often hesitate to rely on computational predictions, especially when burned in the past by models that promised too much and delivered too little.

Carlsson has learned that credibility doesn’t come from publishing a paper or showing a binding score. It comes from honesty. His group is trained to explain exactly what a model can predict and where its limits are. If docking can suggest binding modes but can’t resolve nanomolar potency differences, they say so. If a simulation narrows options but doesn’t rank them definitively, they make that clear.

That candor changes the tone of collaboration. Pharmacologists know they aren’t being oversold. Chemists understand why certain molecules are worth the effort to synthesize. And experiments are better designed because expectations are realistic. Over time, that transparency has built enduring partnerships, including one of Carlsson’s earliest with Ken Jacobson’s lab at the NIH, which continues to advance GPCR ligand discovery today.

Beyond Academia: Advising Biotech

Carlsson’s collaborative philosophy has also found a place in the biotech industry. Through his consulting company, he doesn’t simply sell computational services. Instead, he acts as an advisor, helping research teams decide where modeling can accelerate progress—and where it can’t.

This distinction is critical. Many startups chase elaborate simulations that look impressive on paper but do little to move drug discovery forward. Carlsson helps teams avoid that trap by focusing on leverage points: which models can be trusted for a particular GPCR, which predictions justify immediate synthesis, and when the smartest move is to pause until better biological data arrives.

In practice, this approach saves companies time, resources, and credibility with investors. It also allows Carlsson to stay plugged into the fast-changing needs of biotech while keeping his academic mission intact: training the next generation of GPCR scientists.

The GPCR Challenge

What makes collaboration non-negotiable in GPCR research is the biology itself. GPCRs regulate essential functions from vision to mood to cardiovascular control. They offer extraordinary therapeutic promise—but their complexity makes them one of the hardest targets in drug discovery. Biased signaling, allosteric binding sites, and subtype selectivity mean there is rarely a straight line from prediction to medicine.

No single discipline can cover all the ground. Structural biologists may capture snapshots of receptor conformations, but lack large-scale screening capacity. Modelers can generate binding hypotheses, but they remain untested until validated experimentally. Medicinal chemists can synthesize molecules, but must choose carefully from infinite possibilities. The path forward depends on knitting these pieces together into a coherent discovery engine.

Carlsson’s lab is a proof point: collaboration is not just a cultural preference. It is a technical necessity.

Lessons from the Collaboration Blueprint

Carlsson’s approach offers lessons across the spectrum of GPCR science. At the individual level, progress begins with transparency. Overstating the precision of a model or the predictive power of an assay may look persuasive in the short term, but it erodes trust and slows progress in the long term.

Equally important is the ability to learn across disciplines. A pharmacologist who understands docking, or a modeler who knows assay constraints, will collaborate more effectively and make fewer wrong turns. Science advances faster when people invest in fluency beyond their own silo.

Finally, collaboration should be treated as a deliberate strategy, not an afterthought. The most effective discovery programs are structured like Carlsson’s: modeling designed with downstream chemistry in mind, chemistry connected to assays, and assay data cycling back to refine predictions.

Whether in academia or biotech, the future belongs to research groups that can orchestrate ecosystems rather than defend territories.

Why His Model Works

What sets Carlsson apart is not just access to computational power or assay platforms. It is the values he applies to every collaboration. He treats modeling as part of a continuum rather than a separate discipline. He resists hype, focusing instead on predictions collaborators can actually use. And he prioritizes shared purpose, designing each project with the next experiment in mind.

These values have become the glue that holds his partnerships together. They explain why collaborators return, why projects move forward, and why his model of collaboration is increasingly cited as a template for how modern GPCR science should be done.

GPCR Models Don’t Discover Drugs—People Do

In GPCR research, the biggest breakthroughs won’t come from algorithms or crystal structures alone. They will come from how scientists choose to work together.

Carlsson’s lesson is simple but profound: the strongest model in GPCR drug discovery isn’t built on a supercomputer. It’s built on trust between scientists. And for the next generation of researchers and biotech leaders, that is the real collaboration advantage.