In cancer research, tumor metastasis remains the primary cause of poor patient outcomes. Multi-omics approach—integrating genomics (e.g., mutational landscapes), single-cell transcriptomics (e.g., resolving cellular heterogeneity and rare metastatic subclones), spatial transcriptomics (e.g., mapping tumor-immune microenvironment interactions), proteomics (e.g., protein signaling cascades), and metabolomics (e.g., metabolic reprogramming)—provide unprecedented resolution to dissect the molecular drivers of metastasis.

This integration of complex, high-dimensional data, often heterogeneous, presents significant challenges. Artificial Intelligence (AI), particularly deep learning models, has shown immense promise in processing such extensive datasets. By automating feature extraction and pattern recognition, AI enables the integration of multiple omics levels, enhancing tumor classification and prognostic evaluations. This integration is crucial as the landscape of tumor study evolves towards more comprehensive predictive models which are vital for anticipating disease progression.
This Research Topic aims to spotlight recent advancements in AI applications within the realm of tumor multi-omics. We seek to explore how AI-driven approaches can revolutionize the prediction, analysis, and understanding of tumor metastasis. Specific emphasis will be placed on the development and utilization of predictive models that leverage multi-omics data to anticipate and elucidate disease pathways and outcomes.

To gather further insights in the realm of AI-driven multi-omics and tumor metastasis, we welcome articles addressing, but not limited to, the following themes:

>Multi-omics integration for metastasis-specific biomarker discovery:
Identification of key biomarkers via single-cell/spatial transcriptomics and proteomics.

>AI-driven pattern recognition and predictive modeling in metastasis:
Development of deep learning models to forecast metastasis risk, organotropism (e.g., liver/bone), and temporal progression.

>Mechanistic insights into metastatic progression through AI:
Decoding molecular drivers of metastasis, including tumor heterogeneity, microenvironment interactions, and immune evasion pathways.

>Personalized therapeutic strategies based on AI-powered multi-omics:
Tailoring interventions using metastasis propensity scores profiles derived from multi-omics integration.

>Spatiotemporal modeling of organ-specific metastasis:
Integrative modeling of multi-omics data (e.g., spatial stromal signals, proteomic remodeling) to predict organotropism and metastatic dormancy.

>Translational validation and clinical challenges in AI applications:
Clinical validation in real-world cohorts (e.g., liquid biopsy) and addressing data harmonization (batch effects, multi-center standardization).

Are you interested in contributing a manuscript to this Research Topic?

Artificial Intelligence in Multi-omics: Advancing Tumor Metastasis Prediction and Mechanism Analysis