Arushi Gupta
@arushigupta.bsky.social
Incoming CS PhD at Stanford | Prev. at Caltech
This work began from a summer internship at @msftresearch.bsky.social New England. Incredibly thankful to mentors @alexijie.bsky.social and Alan Moses, whose deep involvement and guidance made this possible!
📄: www.biorxiv.org/content/10.1...
💻: tinyurl.com/microscopy-c...
📄: www.biorxiv.org/content/10.1...
💻: tinyurl.com/microscopy-c...
Representation Learning Methods for Single-Cell Microscopy are Confounded by Background Cells
Deep learning models are widely used to extract feature representations from microscopy images. While these models are used for single-cell analyses, such as studying single-cell heterogeneity, they t...
www.biorxiv.org
July 4, 2025 at 9:50 PM
This work began from a summer internship at @msftresearch.bsky.social New England. Incredibly thankful to mentors @alexijie.bsky.social and Alan Moses, whose deep involvement and guidance made this possible!
📄: www.biorxiv.org/content/10.1...
💻: tinyurl.com/microscopy-c...
📄: www.biorxiv.org/content/10.1...
💻: tinyurl.com/microscopy-c...
Ultimately, our results reveal a key limitation of current crop-based representation learning models for single-cell analysis, and underscore the need for methods that learn background-invariant single-cell features.
July 4, 2025 at 9:50 PM
Ultimately, our results reveal a key limitation of current crop-based representation learning models for single-cell analysis, and underscore the need for methods that learn background-invariant single-cell features.
The background sensitivity of models has consequences for reliable single-cell analysis.
When background cells are masked, predicted localization distributions for multi-localized proteins, or proteins where localization changes from cell to cell, differ substantially for ~15% of proteins.
When background cells are masked, predicted localization distributions for multi-localized proteins, or proteins where localization changes from cell to cell, differ substantially for ~15% of proteins.
July 4, 2025 at 9:50 PM
The background sensitivity of models has consequences for reliable single-cell analysis.
When background cells are masked, predicted localization distributions for multi-localized proteins, or proteins where localization changes from cell to cell, differ substantially for ~15% of proteins.
When background cells are masked, predicted localization distributions for multi-localized proteins, or proteins where localization changes from cell to cell, differ substantially for ~15% of proteins.
We find that three leading single-cell models—PIFiA, Paired Cell Inpainting, and DeepLoc—are sensitive to background context.
Classification accuracy drops by up to 15.8% when background cell localization differs from the center cell, and improves when they match.
Classification accuracy drops by up to 15.8% when background cell localization differs from the center cell, and improves when they match.
July 4, 2025 at 9:50 PM
We find that three leading single-cell models—PIFiA, Paired Cell Inpainting, and DeepLoc—are sensitive to background context.
Classification accuracy drops by up to 15.8% when background cell localization differs from the center cell, and improves when they match.
Classification accuracy drops by up to 15.8% when background cell localization differs from the center cell, and improves when they match.
To assess if background alters model behavior, we evaluate five settings where the center cell is fixed while the background is either unaltered, swapped with backgrounds from the same or different protein localization classes, replaced with a cell-free background, or masked out.
July 4, 2025 at 9:50 PM
To assess if background alters model behavior, we evaluate five settings where the center cell is fixed while the background is either unaltered, swapped with backgrounds from the same or different protein localization classes, replaced with a cell-free background, or masked out.
We focus on representation learning models trained to capture protein localization, the subcellular component within the cell where a protein resides, in yeast cells.
July 4, 2025 at 9:50 PM
We focus on representation learning models trained to capture protein localization, the subcellular component within the cell where a protein resides, in yeast cells.
This matters because understanding single-cell variability is key in biology. If models are altering their behavior depending on background, they are not robust for biological analyses. With @alexijie.bsky.social and Alan Moses, we introduce a framework to systematically test for this.
July 4, 2025 at 9:50 PM
This matters because understanding single-cell variability is key in biology. If models are altering their behavior depending on background, they are not robust for biological analyses. With @alexijie.bsky.social and Alan Moses, we introduce a framework to systematically test for this.