Bryce Allen

CEO @dxtherapeutics; AI/Physics for Drug Discovery; Prev: Co-founder @silicontx (acquired @RoivantDiscover)

[Bryce Allen @bryceallen]

At long last, we describe the phenomenal effort and execution of the team we assembled at @silicontx (acquired by @RoivantDiscover) to leverage integrated computational approaches to difficult biological problems with translational impact.
https://t.co/fAb5N1E487

[Bryce Allen @bryceallen]

It was an honor and a privilege to chair the AI/Machine Learning for Early Discovery and Degraders & Molecular Glues sections of the 2026 Drug Discovery Chemistry Conference.

These sessions brought together some of the brightest minds working at the intersection of computation and small molecule drug discovery, and the quality of science on display was genuinely inspiring. My sincere thanks to all of the speakers who contributed their time and insights, and to the conference organizers for putting together such a thoughtfully curated program.

I also had the opportunity to present our own work in a talk titled "Prediction of Molecular Glues for Challenging Targets in Oncology", sharing how we're applying AI-driven approaches at Differentiated Therapeutics to tackle historically undruggable target classes in cancer biology.

Grateful to be part of a community pushing the boundaries of what's possible in drug discovery.

#DrugDiscoveryChemistry #MolecularGlues #ArtificialIntelligence #Oncology #Degraders #DDC2026

[Bryce Allen @bryceallen]

@vijaypande 🙏 Legend!

[Bryce Allen @bryceallen]

The path from data to discovery is rarely linear.
But in drug discovery, every predictive failure holds a hidden insight.
Strategic iteration isn't just an approach—it's how breakthroughs are born.

[Bryce Allen @bryceallen]

Predictive modeling in drug discovery is only as good as the data it’s built on. High-quality, well-curated datasets can be the difference between uncovering a breakthrough and chasing a mirage. Invest in your foundation—science demands it.

[Bryce Allen @bryceallen]

Research is the heartbeat of biotech, but strategic focus is what turns potential into progress.
Driving molecules toward the clinic isn’t just about data—it's about vision, perseverance, and finding the signal in the noise.

[Bryce Allen @bryceallen]

Drug discovery thrives at the intersection of perseverance and precision.
The path to transformative therapies isn’t linear—it’s a web of insights, failures, and breakthroughs.
Keep building the data, refining the models, and moving molecules forward.

Progress compounds over time.

[Bryce Allen @bryceallen]

Biotech innovation isn’t just about the next molecule—it’s about building systems that get drugs to patients faster.
Focus on platforms that learn, adapt, and scale.
Because the breakthrough is only as impactful as your ability to deliver it to the clinic.

[Bryce Allen @bryceallen]

Drug discovery thrives on the balance between intuition and data-driven insights.
When predictive models align with experimental results, you're not just iterating—you’re accelerating innovation.

The future of medicine belongs to those who find that equilibrium.

[Bryce Allen @bryceallen]

Breakthroughs aren’t just born in the lab—they’re shaped by strategic execution.
The real advantage in biotech?
Building systems that systematically translate great science into clinical impact.

Startups that master this balance leave a lasting legacy.

[Bryce Allen @bryceallen]

Imagine a world where AI and biochemistry work hand in hand to create breakthroughs in medicine.

Curious about what that looks like?

Let's dive in.

Most people assume the future is just faster screening and better prediction engines. But the truth is, we haven't even scratched the surface when we bring human intuition in biochemistry together with deep learning models that are willing to abandon our legacy knowledge.

For example:
Everyone cheers about virtual screening, but what if the true winners are the AI systems that spot non-obvious allosteric sites—sites traditional screens overlook? An AI can suggest pockets for intervention that a medicinal chemist might ignore, unlocking “undruggable” targets.

Here’s the uncomfortable part: the biggest leaps won’t come from improving accuracy by a few percent. They'll come from realizing entire classes of targets and mechanisms have been ignored because our old methods couldn’t see them.

Case in point: Consider intrinsically disordered proteins. Old workflows treat them as noise. AI-driven models are starting to reveal their cryptic binding potential—something classic biochemistry dismissed, but which could unlock therapies for neurodegenerative diseases.

We must stop treating computational chemistry and experimental biology as silos. True innovation happens at their intersection—and it doesn’t always play by the rules we’re used to.

For instance, at Differentiated Therapeutics we've seen hybrid wet-lab/AI closed-loop systems designing, testing, and iterating on molecules far outside traditional chemical space. The result? Structures a seasoned chemist would never dream up—but that actually work, because the AI sees patterns hidden from human intuition.

It won't be easy. It won't be risk-free. But disruption never is.

Who’s ready to challenge consensus and reshape the drug discovery process from the ground up? Let’s connect.

[Bryce Allen @bryceallen]

Imagine a world where AI and biochemistry work hand in hand to create breakthroughs in medicine.

Curious about what that looks like?

Let's dive in.

Most people assume the future is just faster screening and better prediction engines. But the truth is, we haven't even scratched the surface when we bring human intuition in biochemistry together with deep learning models that are willing to abandon our legacy knowledge.

For example:
Everyone cheers about virtual screening, but what if the true winners are the AI systems that spot non-obvious allosteric sites—sites traditional screens overlook? An AI can suggest pockets for intervention that a medicinal chemist might ignore, unlocking “undruggable” targets.

Here’s the uncomfortable part: the biggest leaps won’t come from improving accuracy by a few percent. They'll come from realizing entire classes of targets and mechanisms have been ignored because our old methods couldn’t see them.

Case in point: Consider intrinsically disordered proteins. Old workflows treat them as noise. AI-driven models are starting to reveal their cryptic binding potential—something classic biochemistry dismissed, but which could unlock therapies for neurodegenerative diseases.

We must stop treating computational chemistry and experimental biology as silos. True innovation happens at their intersection—and it doesn’t always play by the rules we’re used to.

For instance, at Differentiated Therapeutics we've seen hybrid wet-lab/AI closed-loop systems designing, testing, and iterating on molecules far outside traditional chemical space. The result? Structures a seasoned chemist would never dream up—but that actually work, because the AI sees patterns hidden from human intuition.

It won't be easy. It won't be risk-free. But disruption never is.

Who’s ready to challenge consensus and reshape the drug discovery process from the ground up? Let’s connect.

[Bryce Allen @bryceallen]

In science, I’ve learned not to over-quantify challenges as they present themselves. Anticipating solutions without context adds noise. Focusing early on data acquisition yields sharper insights, better decisions, and less analytical clutter.

[Bryce Allen @bryceallen]

Innovation isn’t just about bold new ideas—it’s about execution.
How quickly can you turn your concept into value?
How efficiently can you learn from failures and iterate?

Focus on disciplined, relentless progress towards your milestones. Ideas are common; impact isn’t.

[Bryce Allen @bryceallen]

In biotech, strategic efforts can significantly enhance the impact of your research and innovations over time. By focusing on meaningful connections and consistently pushing molecules towards the clinic, you pave the way for lasting influence and breakthroughs.

[Bryce Allen @bryceallen]

@DudespostingWs Smaller boats

[Bryce Allen @bryceallen]

@marwinsegler Nailed it

[Bryce Allen @bryceallen]

@Optibrium @prof_ajay_jain @biopharmics Congrats @prof_ajay_jain and the team at @Optibrium for getting such a talented duo onboard 💪

[Bryce Allen @bryceallen]

@tiwarylab https://t.co/WWDjoKtPoL best viewed on VHS

[Bryce Allen @bryceallen]

A friendly reminder to never underestimate the power of /dev/shm to alleviate the bottlenecks in your IO heavy HPC workflows

[Bryce Allen @bryceallen]

A very valuable tool if you work with cryoEM data

[Structura Biotechnology @structurabio]

1/ We’re thrilled to announce that 3D Flexible Refinement, a motion-based deep generative model for continuous heterogeneity in #cryoEM structures, is available today in #CryoSPARC v4.1 Beta! ❄️⚡

Read more about v4.1: https://t.co/M3mLGtFPIq