arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2412.04072cs.LG
收藏

面向空间转录组基因表达预测的边界引导学习

Boundary-Guided Learning for Gene Expression Prediction in Spatial Transcriptomics

Space AI, Li Auto(理想汽车空间人工智能部门);College of Software, Jilin University(吉林大学软件学院);Computer Science and Engineering, UNSW(新南威尔士大学计算机科学与工程学院)

Mingcheng Qu, Yuncong Wu, Donglin Di, Anyang Su, Tonghua Su, Yang Song, Lei Fan

AI总结:

针对现有空间转录组基因表达预测方法未考虑细胞结构与微环境相互作用的问题,提出含三分支的BG-TRIPLEX框架,利用病理图像边界特征引导学习,在三个公开数据集上PCC指标优于现有方法。

摘要:

空间转录组(Spatial Transcriptomics,ST)是一项可为基因表达提供空间背景的前沿技术。近年来,基于深度学习的方法已展现出利用ST数据从全视野病理切片(WSI)数据中预测基因表达的能力。现有方法通常借助预训练模型从图像及相邻区域提取特征,再设计方法融合这些信息以生成最终输出,但这类方法往往未能考虑细胞结构相似性、细胞密度以及微环境内的相互作用。本文提出名为BG-TRIPLEX的框架,利用从病理图像中提取的边界信息作为引导特征,增强基于WSI的基因表达预测效果。具体而言,该模型包含三个分支:斑点分支、上下文内分支和全局分支。在斑点分支与上下文内分支中,通过预训练模型提取包括边缘和细胞核特征在内的边界信息,这些边界特征通过多头交叉注意力(Multi-Head Cross-Attention)引导细胞形态学与微环境特征的学习,最终将这些特征与全局特征整合以预测最终输出。研究在三个公开ST数据集上开展了大量实验,结果表明BG-TRIPLEX在皮尔逊相关系数(Pearson Correlation Coefficient, PCC)指标上始终优于现有方法。该方法凸显了边界特征在理解WSI与基因表达间复杂相互作用中的关键作用,为未来研究提供了有前景的方向。

英文摘要:

Spatial transcriptomics (ST) has emerged as an advanced technology that provides spatial context to gene expression. Recently, deep learning-based methods have shown the capability to predict gene expression from WSI data using ST data. Existing approaches typically extract features from images and the neighboring regions using pretrained models, and then develop methods to fuse this information to generate the final output. However, these methods often fail to account for the cellular structure similarity, cellular density and the interactions within the microenvironment. In this paper, we propose a framework named BG-TRIPLEX, which leverages boundary information extracted from pathological images as guiding features to enhance gene expression prediction from WSIs. Specifically, our model consists of three branches: the spot, in-context and global branches. In the spot and in-context branches, boundary information, including edge and nuclei characteristics, is extracted using pretrained models. These boundary features guide the learning of cellular morphology and the characteristics of microenvironment through Multi-Head Cross-Attention. Finally, these features are integrated with global features to predict the final output. Extensive experiments were conducted on three public ST datasets. The results demonstrate that our BG-TRIPLEX consistently outperforms existing methods in terms of Pearson Correlation Coefficient (PCC). This method highlights the crucial role of boundary features in understanding the complex interactions between WSI and gene expression, offering a promising direction for future research.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑