Code
Some open-source projects I have worked on. See my github profile.
Delaunay-Watershed
Delaunay-Watershed-3D is an algorithm designed to reconstruct in 3D a sparse surface mesh representation of the geometry of multicellular structures or nuclei from instance segmentations. It accomplishes this by building multimaterial meshes from segmentation masks. These multimaterial meshes are perfectly suited for storage, geometrical analysis, sharing and visualization of data.
Foambryo
foambryo is a python package developed to infer relative surface tensions and cell pressures from the 3D geometry of cells in foam-like clusters, such as early-embryos, tissues or organoids.
Foambryo 2D
A lightweight 2D adaptation of foambryo.
Optiflow
optiflow is an algorithm designed to reconstruct velocity fields from kymographs with optimal transport.
Optimal transport computes the velocity field of minimal work, which is relevant under a wide array of situations and does not necessitates hard physical assumptions, except for the conservation of the physical quantity that we are measuring across time. This software is designed to offer cell biologists a simple method to obtain flow fields around a membrane contour (taking into account boundary conditions) for free, offering a quantitative measurement for a flow that can only be observed visually otherwise.
meshFT
meshFT is a minimalistic PyTorch-based python library that provides a differentiable fourier transform that compute Fourier transform of triangle meshes in a given box. We support operations both on CPU and GPU. We provide C++/CUDA bindings to compute efficiently the forward and backward passes, to enable differentiable rasterization of triangulated data at scale.
The tiny Surface Evolver
The tiny Surface Evolver is a PyTorch-based minimalist implementation of area-minimization on manifolds, inspired by The Surface Evolver, under 120 lines of code.
Outlines MLX
outlinesmlx is a minimalistic library aimed at adapting the Outlines library for use with the MLX framework.
Outlines provides methods to control the generation of language models to make their output more predictable. Combined with MLX, it enables guided generation with large language models while leveraging Apple Silicon hardware.
Physics of multicellular systems
In 2021/2022, I teached lessons on the physics of multicellular systems at the Master 2 of Fundamental Physics at Ecole Normale Superieure. Here are some notebooks I used to help students make physical simulations of biological phenomenas.