Jophin George Joseph
I built Mechanome, a physics-first inference engine that recovers mechanical forces in cells from protein structure and microscopy.
Core trick
Predicts membrane curvature and actin forces from protein structure with biophysics equations, then runs the same model backward on microscopy images.
Tags
Description
I built Mechanome, a physics-first inference engine that recovers mechanical forces in cells from protein structure and microscopy. It predicts forces forward from a protein's structure and runs the same biophysics backward on microscopy to infer them.
Applied to Influenza A viral entry via clathrin-mediated endocytosis, it finds that successful entry requires the simultaneous action of curvature-generating protein insertion and crowding along with actin forces, with force consolidating a shape curvature has already set. The prediction matches real super-resolution imaging across three cell lines, and holds genetically in influenza entry.
This matters because force usually can't be read from microscopic images. Mechanome gives calibrated answers where the data support them, honest refusals where they don't. Mechanome when connected to a LLM can act as a hypothesis generation and validation tool, bringing us one step closer to agentic mechanobiology. Using mechanome in Claude Science I was able to generate an initial manuscript for unique insight into mechanobiology of endocytosis during viral entry.
README excerpt
# Mechanome **Understanding how cells generate, transmit, and respond to forces.** Mechanome is a growing collection of physical models for cell mechanics. It brings together membrane bending, cortical tension, tissue junctions, molecular adhesion, mechanosensitive channels, and protein structure. We want to connect these models to experiments while keeping a clear record of what was measured, what was inferred, and what remains a hypothesis. ## Why we're building it We can often see a cell change shape more easily than we can measure the forces behind that change. Turning an image into a mechanical explanation requires a model of how forces produce the observed geometry. Work on [force inference in epithelial tissues](https://pubmed.ncbi.nlm.nih.gov/22939902/) shows how this can be done, and why comparison with independent measurements matters. In our membrane models, several combinations of tension, curvature sources, and active forces can produce similar shapes. A good fit alone may leave the mechanism unresolved. The physical connections also cross scales. A protein responds to its mechanical environment, while its activity can change that environment. For example, [purified Piezo1 channels respond to forces in a lipid bilayer](https://pubmed.ncbi.nlm.nih.gov/27829145/), connecting membrane mechanics to a molecular response. Studying each component in isolation can miss the conditions under which it matters. We are building Mechanome to make those connections testab