AI and multi-omics reveal how gut bacteria cause colon cancer

Over the past two decades, high-throughput sequencing has dramatically expanded our understanding of the composition of the gut microbiome. However, most studies only enumerate which bacteria are present in colorectal cancer (CRC) patients but do not explain how the bacteria actually work. The field currently faces major analytical hurdles. Microbiome data are compositional (changes in one microorganism cause obvious changes in all other microorganisms), very sparse (most bacteria are absent in most samples), and high-dimensional (thousands of features exist for a small number of patients). These challenges often result in spurious correlations and irreproducible results. Because of these issues, deeper mechanistic studies of how microorganisms actually interact with their human hosts are urgently needed.

Researchers from the Institute of Gastroenterology, Chinese University of Hong Kong, led by Professor Jun Yu and Dr Yinghong Lu, published a review (DOI: 10.20892/j.issn.2095-3941.2025.0762): Cancer biology and medicine. In this article, we systematically analyze how the gut microbiota and host cells interact with each other across four molecular layers: genome, transcriptome, epigenome, and metabolome, highlighting the computational innovations that enable such multiomics integration.

A notable example is bacteria Escherichia coli carry P.K. island. Colibactin produces colibactin, a genotoxin that creates telltale DNA damage signatures found in more than 12% of colorectal cancer cases, a direct molecular fingerprint of the microorganisms that cause cancer mutations. Another villain, Fusobacterium nucleatumuses the virulence factor FadA to latch onto E-cadherin on host cells and switch on Wnt/β-catenin signaling to promote uncontrolled proliferation. This review also focuses on metabolites. Secondary bile acids, such as deoxycholic acid (DCA) produced by certain bacteria, suppress cytotoxic CD8+ T cells and help tumors evade immune attack.

The authors go beyond individual bugs to address the computational crisis in microbiome research. They explain how synthetic artifacts can disguise correlations and how machine learning techniques (random forests and neural networks like MetaNN) cut through the noise. Emerging technologies such as long-read sequencing (PacBio single-molecule real-time sequencing and Oxford Nanopore Technology) and bacterial single-cell spatial transcriptomics (bacterial multiplex error-robust fluorescence in situ hybridization) are also featured. These tools precisely map which bacterial cells reside next to which host cells within tumors, revealing microniches where bacteria drive inflammation and metastasis.

The authors stated that the gut microbiome is not a passive passenger in colorectal cancer, but an active trigger that rewires host biology from the inside out. “For years, we have been looking at lists of bacteria without understanding their true function within tumors,” the researchers explained. “Now, by combining multi-omics and AI-driven models, we can finally see how specific microorganisms disrupt DNA, silence tumor suppressors, and reprogram the immune system.” They added that the biggest challenge is moving from correlation to causation. But with new tools like organ-on-a-chip systems and gnotobiotic mice, that goal is finally within reach, he added.

Insights lead directly to clinical action. Microbiome signatures could one day be used for early screening for colorectal cancer or to predict who will respond to immunotherapy. Elimination of harmful bacteria – for example, the use of phages against enterotoxigenic substances Bacteroides fragilis – Chemosensitivity may be restored. Conversely, manipulating beneficial bacteria to deliver anticancer substances or restore barrier function provides a viable therapeutic strategy. The authors envision a “digital twin” that integrates a patient’s multi-omics data to predict how dietary changes, prebiotics, or biologic therapies will reshape a patient’s individual gut ecosystem. Such precision microbiome medicine has the potential to revolutionize the prevention, diagnosis, and treatment of colorectal cancer.