Jason W. Rocks

Double descent without digital computation

Sam Dillavou, Jason W. Rocks, Jacob F. Wycoff, Andrea J. Liu et autres

The success of large-scale AI relies on the phenomenon of double descent, where overfitting first increases and then counterintuitively decreases as the network grows relative to the amount of training data and becomes overparameterized. In this overparameterized regime it is easy to …

us (code pays fourni par la source)

Local energetic coupling enhances the expressivity of chemical computation

Marco Tuccio, Jason W. Rocks, Joshua E. Goldford

Living systems compute with chemistry by mapping environmental signals onto specific internal chemical states. Despite recent advances in molecular programming, it remains unclear which physicochemical features control the computational expressivity of chemical systems. Here we inverse-design thermodynamically consistent chemical reaction networks whose …

Horizyn-1 model weights and development dataset: Dual-encoder contrastive learning accelerates enzyme discovery

Jason W. Rocks, Dat Truong, Dmitrij Rappoport, Samuel Maddrell-Mander et autres

Overview This repository provides the official model weights for the Horizyn-1 enzyme discovery models, as well as the open-source development dataset, as described in "Dual-encoder contrastive learning accelerates enzyme discovery." The accompanying code for training, inference, and evaluation is available at: https://github.com/dayhofflabs/horizyn. …

Dual-encoder contrastive learning accelerates enzyme discovery

Jason W. Rocks, Dat P. Truong, Dmitrij Rappoport, Samuel Maddrell-Mander et autres

The ability to engineer enzymes for desired reactions is a cornerstone of modern biotechnology, yet identifying suitable starting proteins remains a critical bottleneck. Although contrastive learning offers a compelling computational approach for enzyme discovery, these models have yet to be implemented at …

Horizyn-1 model weights and development dataset: Dual-encoder contrastive learning accelerates enzyme discovery

Jason W. Rocks, Dat Truong, Dmitrij Rappoport, Samuel Maddrell-Mander et autres

Overview This repository provides the official model weights for the Horizyn-1 enzyme discovery models, as well as the open-source development dataset, as described in "Dual-encoder contrastive learning accelerates enzyme discovery." The accompanying code for training, inference, and evaluation is available at: https://github.com/dayhofflabs/horizyn. …

Ultra-high-throughput mapping of genetic design space

Kshitij Rai, Ronan W. O’Connell, Trenton C. Piepergerdes, Yiduo Wang et autres

us, dk (code pays fourni par la source)

Horizyn-1 Development Dataset: Dual-encoder contrastive learning accelerates enzyme discovery

Jason W. Rocks, Dat Truong, Dmitrij Rappoport, Samuel Maddrell-Mander et autres

⚠️ IMPORTANT NOTICE: SUPERSEDED BY VERSION 2 ⚠️ Please use the newest version of this repository. The development model weights (horizyn-v1.ckpt) included in this specific version (v1) are possibly corrupted. A fixed version of the development model (horizyn_v1_0_dev.ckpt), along with the newly …

Dual-encoder contrastive learning accelerates enzyme discovery

Jason W. Rocks, Dmitrij Rappoport, S. Maddrell-Mander, Daniel A. Martin-Alarcon et autres

The ability to engineer enzymes for desired reactions is a cornerstone of modern biotechnology, yet identifying suitable starting proteins remains a critical bottleneck. Although contrastive learning offers a compelling computational approach for enzyme discovery, these models have yet to be implemented at …

Universal Niche Geometry Governs the Response of Ecosystems to Environmental Perturbations

Akshit Goyal, Jason W. Rocks, Pankaj D Mehta

How ecosystems respond to environmental perturbations is a fundamental question in ecology, made especially challenging due to the strong coupling between species and their environment. Here, we introduce a theoretical framework for calculating the steady-state response of ecosystems to environmental perturbations in …

in, us (code pays fourni par la source)

Engineering synthetic phosphorylation signaling networks in human cells

Xiaoyu Yang, Jason W. Rocks, Kaiyi Jiang, Andrew J. Walters et autres

Protein phosphorylation signaling networks have a central role in how cells sense and respond to their environment. We engineered artificial phosphorylation networks in which reversible enzymatic phosphorylation cycles were assembled from modular protein domain parts and wired together to create synthetic phosphorylation …

us (code pays fourni par la source)

Integrating local energetics into Maxwell-Calladine constraint counting to design mechanical metamaterials

Jason W. Rocks, Pankaj D Mehta

The Maxwell-Calladine index theorem plays a central role in our current understanding of the mechanical rigidity of discrete materials. By considering the geometric constraints each material component imposes on a set of underlying degrees of freedom, the theorem relates the emergence of …

us (code pays fourni par la source)

Phase Transition to Chaos in Complex Ecosystems with Nonreciprocal Species-Resource Interactions

Emmy Blumenthal, Jason W. Rocks, Pankaj D Mehta

Non-reciprocal interactions between microscopic constituents can profoundly shape the large-scale properties of complex systems. Here, we investigate the effects of non-reciprocity in the context of theoretical ecology by analyzing a generalization of MacArthur's consumer-resource model with asymmetric interactions between species and resources. …

us (code pays fourni par la source)