NEST: Spatially-mapped cell-cell communication patterns using a deep learning-based attention mechanism. NEST: Spatially-mapped cell-cell communication patterns using a deep learning-based attention mechanism

Dysregulation of communication between cells mediates complex diseases such as cancer and diabetes. However, detecting cell-cell communication (CCC) at scale remains one of the greatest challenges in transcriptomics. While gene expression remains one of the greatest challenges in transcriptomics. While gene expression measured with single-cell RNA sequencing and spatial transcriptomics reinvigorated computational approaches to detecting CCC, most existing methods exhibit high false positive rates, do not integrate spatial proximity of ligand-receptor interactions, and cannot detect CCC between individual cells. We overcome these challenges by presenting NEST (NEural network on Spatial Transcriptomics), which uses a graph attention network paired with an unsupervised contrastive learning approach to decipher patterns of communication while retaining the strength of each signal. We introduce new synthetic benchmarking experiments which demonstrate how NEST outperforms existing tools and detects biologically-relevant CCC along with directionality and confidence across spot- and cell-based technologies measuring several different tissues and diseases. In our applications, NEST identifies T-cell homing signals in human lymph nodes, aggressive cancer CCC in lung adenocarcinoma, and discovers new patterns of communication in pancreatic cancer. Beyond two-dimensional data, we also highlight NEST's ability to detect CCC in three-dimensional spatial transcriptomic data. Overall design: Two patient samples with resectable (Stage IIB) PDAC were included (Male, 76 years old and Female, 83 years old). Both patient samples were collected from the University Heath Network Biospecimens Program (Toronto, Canada). Ethical approval was obtained through the University Health Network Research Ethics Board (13-6377).

细胞间通讯失调可介导癌症、糖尿病等复杂疾病的发生发展。然而，大规模检测细胞间通讯（cell-cell communication, CCC）仍是转录组学领域的重大挑战之一。尽管借助单细胞RNA测序和空间转录组学获取的基因表达数据为细胞间通讯检测的计算方法注入了新活力，但多数现有方法仍存在诸多局限：假阳性率偏高、未整合配体-受体相互作用的空间邻近性，且无法实现单个细胞间通讯的精准检测。

我们针对上述局限提出了NEST（NEural network on Spatial Transcriptomics，空间转录组学神经网络），该方法结合图注意力网络与无监督对比学习框架，在保留各类信号特征的同时解析细胞间通讯模式。我们设计了全新的合成基准测试实验，证实NEST在多种组织与疾病的基于斑点和单细胞的检测技术中，均优于现有工具，可检测具有生物学意义的细胞间通讯，同时给出通讯方向与置信度评分。

在应用场景中，NEST可在人类淋巴结中识别T细胞归巢信号、在肺腺癌中鉴定侵袭性癌症相关细胞间通讯，并在胰腺癌中发现新的细胞间通讯模式。除二维数据外，我们还验证了NEST在三维空间转录组数据中检测细胞间通讯的能力。

实验整体设计：本研究纳入2例可切除（IIB期）胰腺导管腺癌（pancreatic ductal adenocarcinoma, PDAC）患者样本，分别为76岁男性与83岁女性。所有样本均采集自加拿大多伦多大学健康网络生物标本库项目。本研究已通过大学健康网络研究伦理委员会的伦理审批，审批号为13-6377。