Partnering with Macrocosmos to Accelerate Next-Generation NNP Development

May 1, 2025

Neural network potentials (NNPs) are revolutionizing molecular simulation. Starting today, Rowan is teaming up with Macrocosmos, an open-source AI research lab building on Bittensor, to accelerate the development of the next generation of NNPs through Subnet 25 - Mainframe.

From Macrocosmos' documentation on Subnet 25 - Mainframe:

Mainframe is a decentralised science subnet on Bittensor. It provides computing power and community talent to solve scientific problems.

Subnet 25 currently tackles decentalized protein folding using molecular dynamics (MD) a method for simulating the physical movements of atoms and molecules.

Our collaboration brings Bittensor closer to cutting-edge research in drug discovery and materials science. We look forward to working with the Macrocosmos team to add new capabilities to Mainframe.

In our newly released preprint "Egret-1: Pretrained Neural Network Potentials for Efficient and Accurate Bioorganic Simulation," we present a new family of open-source ML models for high-accuracy bioorganic simulations. One key insight from the paper is that the next generation of models will need a lot more high-quality data generated via density-functional theory (DFT). From the paper:

In particular, we anticipate that a combination of improved dataset scale and quality, more expressive architectures, and performance optimization will make it possible to achieve significantly improved accuracy, speed, and generality, which we expect to have a substantial impact on discovery across the chemical sciences.

Working with Macrocosmos and Mainframe will let us use the decentralized computing power of the Bittensor network to dynamically generate the data we need to train the next generation of NNPs.

Bittensor is an ideal partner for this ambitious scientific project. We're very excited to be working with Macrocosmos, Subnet 25 - Mainframe, and the community to combine our expertise and push the boundaries of ML-powered molecular simulation forward.

Banner background image

Ready to learn more?

Talk with a member of our team to learn how Rowan can accelerate your work. Security and compliance questions can go to security@rowansci.com, and our full security page covers current practices.

Contact us →

What to read next

Benchmarking Membrane-Permeability Predictors

Benchmarking Membrane-Permeability Predictors

Testing GNN-MTL and PyPermm on datasets of small molecules, macrocycles, and PROTACs
Apr 28, 2026 · Ari Wagen
Smarter Analogue Docking, Pocket Detection, and g-xTB Analytical Gradients

Smarter Analogue Docking, Pocket Detection, and g-xTB Analytical Gradients

more robust MCS detection; conformer sampling with torsional Monte Carlo; better alignment and RBFE results; a new pocket-detection workflow; analytical gradients now available for g-xTB
Apr 23, 2026 · Zachary Fried, Corin Wagen, Ari Wagen, and Jonathon Vandezande
g-xTB pKa and Website Redesign

g-xTB pKa and Website Redesign

the flaws with Rowan's AIMNet2-based pKa method; our new g-xTB-based approach; benchmarking and availability; a logo and new website for Rowan
Apr 15, 2026 · Corin Wagen and Ari Wagen
Easter Updates to Rowan

Easter Updates to Rowan

webhooks, draft workflows, and usage estimates for Rowan's Python API; tautomers in non-aqueous solvents; COSMO-based descriptors; overage-based billing; an FEP speed test; welcome Zach
Apr 9, 2026 · Eli Mann, Ari Wagen, Spencer Schneider, Jonathon Vandezande, and Corin Wagen
How Fast Can FEP Run?

How Fast Can FEP Run?

Pushing the speed limit for RBFE calculations run through TMD.
Apr 8, 2026 · Corin Wagen
Improving Rowan's API

Improving Rowan's API

API as a coequal interface to Rowan's product; what we're changing in v3.0.0 of rowan-python; typed outputs; new workflow API; more agent-friendly features; acknowledging our early partners here
Mar 19, 2026 · Eli Mann, Corin Wagen, Jonathon Vandezande, and Spencer Schneider
Building Modern AI-Enabled Infrastructure for Pharma: A Conversation with Anthony Bradley from Dalton

Building Modern AI-Enabled Infrastructure for Pharma: A Conversation with Anthony Bradley from Dalton

Corin talks with Anthony about the real problems in computer-assisted drug discovery, how to sell software to pharma, and what Dalton can learn from Nike.
Mar 17, 2026 · Corin Wagen
Free-Energy Perturbation

Free-Energy Perturbation

what FEP is and why it's useful; limitations of current methods; Rowan FEP, TMD, and public benchmarks; how to run FEP in Rowan; the dream of FEP "too cheap to meter"; how to try Rowan FEP
Mar 4, 2026 · Corin Wagen, Eli Mann, Ari Wagen, and Spencer Schenider
Free-Energy Perturbation: A Pedagogical Introduction

Free-Energy Perturbation: A Pedagogical Introduction

Learn the core concepts behind free energy perturbation (FEP) using interactive 1D toy systems with exact analytical results.
Mar 4, 2026 · Corin Wagen
Solvent-Dependent Conformer Search

Solvent-Dependent Conformer Search

a good conformer is hard to find; clustering and the ReSCoSS workflow; Rowan's implementation, with some expert help; a demonstration on maraviroc
Feb 26, 2026 · Corin Wagen and Ari Wagen