Evaluating Deep Regression Models for WSI-Based Gene-Expression Prediction

Fredrik K. Gustafsson, Mattias Rantalainen
10/1/2024
q-bio.GNcs.CVcs.LG

No official code found. Best implementation: mahmoodlab/UNI (758 stars, last commit 2025-03-26).

Abstract

Prediction of mRNA gene-expression profiles directly from routine whole-slide images (WSIs) using deep learning models could potentially offer cost-effective and widely accessible molecular phenotyping. While such WSI-based gene-expression prediction models have recently emerged within computational pathology, the high-dimensional nature of the corresponding regression problem offers numerous design choices which remain to be analyzed in detail. This study provides recommendations on how deep regression models should be trained for WSI-based gene-expression prediction. For example, we conclude that training a single model to simultaneously regress all 20530 genes is a computationally efficient yet very strong baseline.

View on arXivView PDF

Code Implementations(1)

mahmoodlab/UNICommunity100%

Pathology Foundation Model - Nature Medicine

75890Python, Jupyter NotebookFeb 2, 20241 years agoNOASSERTION
computational-pathologydigital-pathologyfoundationfoundation-modelhistopathology+12 more

Similar Papers

Completing Spatial Transcriptomics Data for Gene Expression Prediction Benchmarking

84% match2025

GeneQuery: A General QA-based Framework for Spatial Gene Expression Predictions from Histology Images

83% match1 repo32024

Self-supervised learning on gene expression data

82% match2025

MMAP: A Multi-Magnification and Prototype-Aware Architecture for Predicting Spatial Gene Expression

82% match2025

Spatial Transcriptomics Analysis of Zero-shot Gene Expression Prediction

82% match2024

Accurate Spatial Gene Expression Prediction by Integrating Multi-Resolution Features

81% match1 repo522026

Cite this paper

@article{gustafsson2024evaluating,
  title  = {Evaluating Deep Regression Models for WSI-Based Gene-Expression Prediction},
  author = {Fredrik K. Gustafsson and Mattias Rantalainen},
  year   = {2024},
  eprint = {2410.00945},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2410.00945v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2410.00945.svg)](https://codeofpaper.com/paper/2410.00945)

Discussion