Marius Pachitariu

Featured in: Favicon

nature.com Favicon

plos.org

science.org Favicon

cell.com

biorxiv.org

Articles

A critical initialization for biological neural networks

Jan 10, 2025 | biorxiv.org | Marius Pachitariu |Lin Zhong |Alexa Gracias |Amanda Minisi

AbstractArtificial neural networks learn faster if they are initialized well. Good initializations can generate high-dimensional macroscopic dynamics with long timescales. It is not known if biological neural networks have similar properties. Here we show that the eigenvalue spectrum and dynamical properties of large-scale neural recordings in mice (two-photon and electrophysiology) are similar to those produced by linear dynamics governed by a random symmetric matrix that is critically normalized.
Analysis methods for large-scale neuronal recordings

Nov 7, 2024 | science.org | Gordon Rix |Julia Phan |Carsen Stringer |Marius Pachitariu

Editor’s summaryRecent years have seen the development of a substantial number of new approaches to investigating the brain. Some of the most promising are techniques that allow functional recordings of large numbers of neurons with electrophysiological or optical imaging methods.
Bridging tuning and invariance with equivariant neuronal representations

Aug 6, 2024 | biorxiv.org | Lin Zhong |Marius Pachitariu |Sandro Romani |Judith Hoeller

AbstractAs we move through the world, we see the same visual scenes from different perspectives. Although we experience perspective deformations, our perception of a scene remains stable. This raises the question of which neuronal representations in visual brain areas are perspective-tuned and which are invariant.
Spike sorting with Kilosort4 - Nature Methods

Apr 8, 2024 | nature.com | Marius Pachitariu |Shashwat Sridhar

AbstractSpike sorting is the computational process of extracting the firing times of single neurons from recordings of local electrical fields. This is an important but hard problem in neuroscience, made complicated by the nonstationarity of the recordings and the dense overlap in electrical fields between nearby neurons. To address the spike-sorting problem, we have been openly developing the Kilosort framework.
Transformers do not outperform Cellpose

Apr 7, 2024 | biorxiv.org | Carsen Stringer |Marius Pachitariu

AbstractIn a recent publication, Ma et al (2024) claim that a transformer-based cellular segmentation method called Mediar - which won a Neurips challenge - outperforms Cellpose (0.897 vs 0.543 median F1 score). Here we show that this result was obtained by artificially impairing Cellpose in multiple ways. When we removed these impairments, Cellpose outperformed Mediar (0.861 vs 0.826 median F1 score on the updated test set).
Cellpose3: one-click image restoration for improved cellular segmentation

Feb 12, 2024 | biorxiv.org | Carsen Stringer |Marius Pachitariu

AbstractGeneralist methods for cellular segmentation have good out-of-the-box performance on a variety of image types. However, existing methods struggle for images that are degraded by noise, blurred or undersampled, all of which are common in microscopy. We focused the development of Cellpose3 on addressing these cases, and here we demonstrate substantial out-of-the-box gains in segmentation and image quality for noisy, blurry or undersampled images.
Facemap: a framework for modeling neural activity based on orofacial tracking - Nature Neuroscience

Nov 20, 2023 | nature.com | Atika Syeda |Lin Zhong |Renee Tung |Marius Pachitariu

AbstractRecent studies in mice have shown that orofacial behaviors drive a large fraction of neural activity across the brain. To understand the nature and function of these signals, we need better computational models to characterize the behaviors and relate them to neural activity. Here we developed Facemap, a framework consisting of a keypoint tracker and a deep neural network encoder for predicting neural activity.
Reading the mouse mind from its face: New tool decodes neural activity using facial movements

Nov 20, 2023 | janelia.org | Lin Zhong |Renee Tung |Will Long |Marius Pachitariu

Mice are always in motion. Even if there’s no external motivation for their actions—like a cat lurking a few feet away—mice are constantly sweeping their whiskers back and forth, sniffing around their environment and grooming themselves. These spontaneous actions light up neurons across many different regions of the brain, providing a neural representation of what the animal is doing moment-by-moment across the brain. But how the brain uses these persistent, widespread signals remains a mystery.