ModelMole

We combine cutting-edge computational chemistry with advanced ML/AI algorithms to identify novel therapeutic targets.

14/05/2026

While it can be difficult to directly compare technologies we can get decent estimates in the baseline unit of currency in this business from what machines different orgs use and for how long they use them.

Here we see how a traditional bruteforce screening approach uses huge amounts of energy, and while traditional AI approaches are better, ModelMole was able to blaze past both in efficiency using its Search in the Dark technology to achieve the same results at 50x less energy usage!

10/03/2026

Lucky to be seated next to at the RSC Investment Catalyst Event. Great conversation and a very productive discussion.

22/01/2026

A year ago i asked an LLM to convert some of my python to CUDA, it took 5 hours and a lot of manual intervention.

Today I asked to convert some python to multithreaded rust and it did the whole thing in 1 shot in under 10 minutes.

02/01/2026

🌟 Key Highlights of 2025:
✅ Fueling Innovation: We were honored to receive a £90k grant from the UK Government, recognizing our mission and potential to drive change in AI directed drug discovery.
✅ Scaling Up: With the support of significant private-sector investment, we’re now better equipped to accelerate our vision and bring even more value to our partners.
✅ Global Reach: This year, we welcomed new clients across the Nordics, EU, and China.
✅ Team Growth: Our team expanded its capacity by transitioning all members to full-time roles.

30/12/2025

Both rely on iterative improvement:
In quantum chemistry, we start with a guess for orbitals or electron density, then update them until the solution converges (the SCF loop).

In AI, we start with random weights and iteratively adjust them to minimize loss.

It’s the same mindset: make a guess, measure error, refine, and repeat until you reach a consistent, useful solution.

And here’s where it gets powerful: DFT isn’t just like AI, it fuels AI. High-quality DFT calculations generate reliable data on molecular energies, reaction barriers, and material properties.

That data trains machine learning models to predict real-world behavior without running expensive simulations every time, from drug discovery to battery design. So while DFT and AI both “learn” through iteration, together they’re accelerating science faster than either could alone.

15/12/2025

We’re still buzzing from the incredible experience at the RSC Change Makers 2025 end-of-year event. A huge thank you to the Royal Society of Chemistry for inviting ModelMole. It is an honour to be part of such a dynamic and forward-thinking community.

Every conversation we had only reinforced how transformative the Change Makers program is. The passion, innovation, and impact we witnessed were truly inspiring, and we left feeling even more motivated to contribute.

We’re thrilled to announce that ModelMole will be part of the 2026 cohort!

13/11/2025

We are proud to announce that ModelMole has been invited into the Royal Society of Chemistry’s (RSC) 2026 Change Makers cohort!

06/11/2025

Tooth enamel – the hardest tissue in the human body – has always been a challenge to restore. Once lost, it doesn’t regenerate naturally. But what if we could rebuild it?

A groundbreaking study published in Nature Communications introduces a biomimetic supramolecular protein matrix that mimics the natural enamel-developing process. Using elastin-like recombinamers (ELRs), researchers engineered a matrix that triggers the epitaxial growth of apatite nanocrystals, recreating enamel’s intricate microstructure and restoring its mechanical properties – stiffness, hardness, and wear resistance.

How?
- Our lead scientist (Alexander van Teijlingen) led the computational work, revealing how ELR molecules self-assemble into fibrils in the presence of calcium ions, forming a scaffold for mineralisation.
- The matrix was applied to eroded teeth, where it guided the growth of organised, enamel-like layers – even on exposed dentine.
- Mechanical tests confirmed the restored enamel matched native tissue in strength and durability, resisting abrasion, acid erosion, and fracture.

Why it matters:
This technology could revolutionise dental care, offering a clinically friendly solution for enamel erosion, hypersensitivity, and decay. By combining biomimicry and computational modelling, we’re closer than ever to regenerating enamel – not just repairing it.

🔬 Science meets innovation: The future of dental health is here.
https://www.nature.com/articles/s41467-025-64982-y

28/10/2025

When Bacteria Win, We All Lose. Here's How We're Fighting Back.

Imagine a world where routine surgeries become life-threatening. Where pneumonia is untreatable. Where a simple cut could kill you.

That's the world antimicrobial resistance (AMR) is creating. Drug-resistant infections already kill over 1 million people every year.

Traditional antibiotics are failing because bacteria evolve resistance to them. We need weapons bacteria can't easily adapt to.

Enter: membrane-disrupting peptides.

Instead of targeting a specific bacterial protein (which bacteria can mutate), these peptides physically punch holes in bacterial cell membranes. It's like the difference between picking a lock and smashing down the door—much harder to defend against.

But here's the challenge: there are trillions of possible peptide sequences. Testing them all in the lab would take centuries.

So we built an AI to do it in months.

This is how we'll stay ahead of resistance: by exploring chemical space faster than bacteria can evolve.

Full paper: https://doi.org/10.1039/D4CP01404A

21/10/2025

Proline is famous for putting kinks in helices - but what if we use its rigidity as a strategic lock?

In this new preprint demonstrating how N-terminal proline constrains backbone dynamics in tripeptides, the authors were able to isolate side-chain effects on assembly.

They looked at PXX (X+F/W) peptides and found that tryptophan's diverse interactions (π-π, N-H...π, H-bonding) create shallow energy landscapes with multiple accessible states. Phenylalanine's limited interaction space locks peptides into fewer conformations.

The locked-backbone strategy elegantly decouples variables to derive sequence-dynamics design rules - from molecular conformations to crystallisation to bulk mechanical properties.

Sometimes holding something still is the best way to understand what moves.

doi.org/10.26434/chemrxiv-2025-q6wqd

Address

Lynwood House, Crofton Road
Orpington
BR68QE

Website

https://modelmole.com/

Alerts

Be the first to know and let us send you an email when ModelMole posts news and promotions. Your email address will not be used for any other purpose, and you can unsubscribe at any time.

Shortcuts

Want your business to be the top-listed Engineering Company in Orpington?