Research

All embodied agents are fundamentally patterns in physiological or other excitable media, blurring the distinction between objects and processes. What competencies do these patterns possess? We equip the creatures from Chan's Lenia with the capacity to sense regions of occlusion, via a biologically inspired modification to the update rule. These occlusions act as obstacles and are received in a variety of ways. When not immediately destroyed, many creatures steer away from occlusions — some immediately, some only after turning toward the obstacle. Occlusions can serve to push a creature into a new dynamical regime, changing its morphology and by proxy, its identity. Occasionally, occlusions provoke the generation of a second, identical creature. Response depends jointly on the occlusion's size, shape, and placement relative to the creature's heading and on the parameters of the creature itself. We map four test creatures' sensitivity to targeted occlusions and interpret the results in the language of dynamical systems. We observe Lenia creatures taking advantage of their freedom to change heading in order to achieve what appears to be a more fundamental goal: the preservation of their morphology. This work illustrates the beginning of an important roadmap to understand how emergent agents' behavioral propensities interact with the informational, not only tangible, topography of their world.

Recent advancements in protein structure prediction methods have vastly increased the size of databases of protein structures, necessitating fast methods for protein structure comparison. Search methods that find structurally similar proteins can be applied to find remote homologs, study the functional relationships among proteins, and aid in protein engineering tasks. We design a "3Dn" structural alphabet that encodes the local neighborhoods around each amino acid in an interpretable way. In a search benchmark task, a combination of our alphabet and Foldseek's 3Di alphabet, outperforms each alphabet individually and ranks best among local search methods that do not require amino acid identity information. We provide software tools that enable the exploration of novel alphabets and combinations of alphabets for protein structure search.