Decoding cancer prognosis with deep learning: the ASD-cancer framework for tumor microenvironment analysis DOI Creative Commons
Ziyuan Huang,

Yunzhan Li,

Vanni Bucci

et al.

mSystems, Journal Year: 2025, Volume and Issue: unknown

Published: April 16, 2025

ABSTRACT Deep learning is revolutionizing biomedical research by facilitating the integration of multi-omics data sets while bridging classical bioinformatics with existing knowledge. Building on this powerful potential, Zhang et al. proposed a semi-supervised framework called Autoencoder-Based Subtypes Detector for Cancer (ASD-cancer) to improve analysis (H. Zhang, X. Xiong, M. Cheng, al., 2024, mSystems 9:e01395-24, https://doi.org/10.1128/msystems.01395-24 ). By utilizing autoencoders pre-trained The Genome Atlas data, ASD-cancer outperforms baseline model. This approach also makes scalable, enabling it process new through transfer without retraining. commentary explores methodological innovations and scalability suggesting future directions, such as incorporation additional layers development adaptive AI models continuous learning. Notably, integrating large language into could enhance its interpretability, providing more profound insights oncological increasing influence in cancer subtyping further analysis.

Language: Английский

Decoding cancer prognosis with deep learning: the ASD-cancer framework for tumor microenvironment analysis DOI Creative Commons
Ziyuan Huang,

Yunzhan Li,

Vanni Bucci

et al.

mSystems, Journal Year: 2025, Volume and Issue: unknown

Published: April 16, 2025

ABSTRACT Deep learning is revolutionizing biomedical research by facilitating the integration of multi-omics data sets while bridging classical bioinformatics with existing knowledge. Building on this powerful potential, Zhang et al. proposed a semi-supervised framework called Autoencoder-Based Subtypes Detector for Cancer (ASD-cancer) to improve analysis (H. Zhang, X. Xiong, M. Cheng, al., 2024, mSystems 9:e01395-24, https://doi.org/10.1128/msystems.01395-24 ). By utilizing autoencoders pre-trained The Genome Atlas data, ASD-cancer outperforms baseline model. This approach also makes scalable, enabling it process new through transfer without retraining. commentary explores methodological innovations and scalability suggesting future directions, such as incorporation additional layers development adaptive AI models continuous learning. Notably, integrating large language into could enhance its interpretability, providing more profound insights oncological increasing influence in cancer subtyping further analysis.

Language: Английский

Citations

0