SpaRED benchmark: Enhancing Gene Expression Prediction from Histology Images with Spatial Transcriptomics Completion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mejia, Gabriel, Ruiz, Daniela, Cárdenas, Paula, Manrique, Leonardo, Vega, Daniela, Arbeláez, Pablo
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916412659859456
author Mejia, Gabriel
Ruiz, Daniela
Cárdenas, Paula
Manrique, Leonardo
Vega, Daniela
Arbeláez, Pablo
author_facet Mejia, Gabriel
Ruiz, Daniela
Cárdenas, Paula
Manrique, Leonardo
Vega, Daniela
Arbeláez, Pablo
contents Spatial Transcriptomics is a novel technology that aligns histology images with spatially resolved gene expression profiles. Although groundbreaking, it struggles with gene capture yielding high corruption in acquired data. Given potential applications, recent efforts have focused on predicting transcriptomic profiles solely from histology images. However, differences in databases, preprocessing techniques, and training hyperparameters hinder a fair comparison between methods. To address these challenges, we present a systematically curated and processed database collected from 26 public sources, representing an 8.6-fold increase compared to previous works. Additionally, we propose a state-of-the-art transformer based completion technique for inferring missing gene expression, which significantly boosts the performance of transcriptomic profile predictions across all datasets. Altogether, our contributions constitute the most comprehensive benchmark of gene expression prediction from histology images to date and a stepping stone for future research on spatial transcriptomics.
format Preprint
id arxiv_https___arxiv_org_abs_2407_13027
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SpaRED benchmark: Enhancing Gene Expression Prediction from Histology Images with Spatial Transcriptomics Completion
Mejia, Gabriel
Ruiz, Daniela
Cárdenas, Paula
Manrique, Leonardo
Vega, Daniela
Arbeláez, Pablo
Computer Vision and Pattern Recognition
Spatial Transcriptomics is a novel technology that aligns histology images with spatially resolved gene expression profiles. Although groundbreaking, it struggles with gene capture yielding high corruption in acquired data. Given potential applications, recent efforts have focused on predicting transcriptomic profiles solely from histology images. However, differences in databases, preprocessing techniques, and training hyperparameters hinder a fair comparison between methods. To address these challenges, we present a systematically curated and processed database collected from 26 public sources, representing an 8.6-fold increase compared to previous works. Additionally, we propose a state-of-the-art transformer based completion technique for inferring missing gene expression, which significantly boosts the performance of transcriptomic profile predictions across all datasets. Altogether, our contributions constitute the most comprehensive benchmark of gene expression prediction from histology images to date and a stepping stone for future research on spatial transcriptomics.
title SpaRED benchmark: Enhancing Gene Expression Prediction from Histology Images with Spatial Transcriptomics Completion
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.13027