CytoSummaryNet is a Deep Sets-based approach that uses self-supervised contrastive learning in a multiple-instance learning framework. Try it out!
Paper: doi.org/10.1371/jour...
Code: github.com/carpenter-si...
With @johnarevalo.bsky.social @drannecarpenter.bsky.social Mehrtash Babadi
3/n
![(a) Human U2OS cells treated with dimethyl sulfoxide (DMSO) and stained using the Cell Painting assay, which employs six dyes in five channels to label eight cellular compartments. The top row (from left to right) shows mitochondrial staining; actin, Golgi, and plasma membrane staining; and nucleolar and cytoplasmic RNA staining. The bottom row (from left to right) displays endoplasmic reticulum staining, DNA staining, and a montage of all five channels (from Cimini et al. [21]). (b) Thousands of features are extracted from each segmented cell in microscopy images of wells. A learned function f(x) (CytoSummaryNet) aggregates this data into a single feature vector: the sampleβs profile. (c) An in-depth look at the model architecture used in this study. The model consists of three elements: a function Ο(x), which maps the input data from βD to βL space, a summation, which collapses the cell dimension, and Ο(z), which maps the collapsed representation from βN to βL space. (d) During training, replicate compound profiles are forced to attract each other (green arrows) and simultaneously repel every other compound (red arrows) in the learned feature space. Here, all forces are drawn for a single profile of compound B.](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:raswr6mjt7sbppdo6kcce4mz/bafkreicppuq2utmxkx5kpaggndigq7qcat6ppdqqajxhphpyi53ydlbbeq)
(a) Human U2OS cells treated with dimethyl sulfoxide (DMSO) and stained using the Cell Painting assay, which employs six dyes in five channels to label eight cellular compartments. The top row (from left to right) shows mitochondrial staining; actin, Golgi, and plasma membrane staining; and nucleolar and cytoplasmic RNA staining. The bottom row (from left to right) displays endoplasmic reticulum staining, DNA staining, and a montage of all five channels (from Cimini et al. [21]). (b) Thousands of features are extracted from each segmented cell in microscopy images of wells. A learned function f(x) (CytoSummaryNet) aggregates this data into a single feature vector: the sampleβs profile. (c) An in-depth look at the model architecture used in this study. The model consists of three elements: a function Ο(x), which maps the input data from βD to βL space, a summation, which collapses the cell dimension, and Ο(z), which maps the collapsed representation from βN to βL space. (d) During training, replicate compound profiles are forced to attract each other (green arrows) and simultaneously repel every other compound (red arrows) in the learned feature space. Here, all forces are drawn for a single profile of compound B.
Taking pictures of cells with a microscope, then extracting thousands of features from them is uncannily effective for quantifying cell state, esp. for genes and chemicals (e.g., Cell Painting). But we often average the rich single-cell data to simplify analysis. Can we do better?
#bioML π§ͺ
1/n
Excited to present our spotlight paper at #NeurIPS!
MOTIVE is a new dataset + benchmark for predicting drug-target interactions, using Cell Painting data
Location: Fri 13 Dec 4:30 p.m. PST @ East Exhibit Hall A-C #4208
Poster: neurips.cc/virtual/2024...
Paper: arxiv.org/abs/2406.08649
