PT  - JOURNAL ARTICLE
AU  - Xiaofan Liu
AU  - Yuhuan Tao
AU  - Zilin Cai
AU  - Pengfei Bao
AU  - Hongli Ma
AU  - Kexing Li
AU  - Yunping Zhu
AU  - Zhi John Lu
TI  - Pathformer: biological pathway informed Transformer model integrating multi-modal data of cancer
AID  - 10.1101/2023.05.23.541554
DP  - 2023 Jan 01
TA  - bioRxiv
PG  - 2023.05.23.541554
4099  - http://biorxiv.org/content/early/2023/05/24/2023.05.23.541554.short
4100  - http://biorxiv.org/content/early/2023/05/24/2023.05.23.541554.full
AB  - Multi-modal biological data integration can provide comprehensive views of gene regulation and cell development. However, conventional integration methods rarely utilize prior biological knowledge and lack interpretability. To address these challenges, we developed Pathformer, a biological pathway informed deep learning model based on Transformer with bias to integrate multi-modal data. Pathformer leverages criss-cross attention mechanism to capture crosstalk between different biological pathways and between different modalities (i.e., multi-omics). It also utilizes SHapley Additive Explanation method to reveal key pathways, genes, and regulatory mechanisms. Through benchmark studies on 28 TCGA datasets, we demonstrated the superior performance and interpretability of Pathformer on various cancer classification tasks, compared to other integration models. Furthermore, we applied Pathformer to liquid biopsy multi-modal data integration with high accuracy in cancer diagnosis. Meanwhile, Pathformer revealed interesting molecularly altered pathways in cancer patients’ body fluid, such as ligand binding of scavenger receptors, iron transport, and DAP12 signaling transmission, which are related to extracellular vesicle transport, platelet, and immune response.Competing Interest StatementThe authors have declared no competing interest.