Artificial intelligence-enabled microbiome-based diagnosis models for a broad spectrum of cancer types

Abstract

Microbiome-based diagnosis of cancer is an increasingly important supplement for the genomics approach in cancer diagnosis, yet current models for microbiome-based diagnosis of cancer face difficulties in generality: not only diagnosis models could not be adapted from one cancer to another, but models built based on microbes from tissues could not be adapted for diagnosis based on microbes from blood. Therefore, a microbiome-based model suitable for a broad spectrum of cancer types is urgently needed. Here we have introduced DeepMicroCancer, a diagnosis model using artificial intelligence techniques for a broad spectrum of cancer types. Built based on the random forest models it has enabled superior performances on more than twenty types of cancers’ tissue samples. And by using the transfer learning techniques, improved accuracies could be obtained, especially for cancer types with only a few samples, which could satisfy the requirement in clinical scenarios. Moreover, transfer learning techniques have enabled high diagnosis accuracy that could also be achieved for blood samples. These results indicated that certain sets of microbes could, if excavated using advanced artificial techniques, reveal the intricate differences among cancers and healthy individuals. Collectively, DeepMicroCancer has provided a new venue for accurate diagnosis of cancer based on tissue and blood materials, which could potentially be used in clinics.

Complexity of human death: its physiological, transcriptomic, and microbiological implications

“…Artificial intelligence (AI) will be assessed in assisting the medical field with processing the massive inundation of data generated from microbiome functional analyses and host-microbiome interaction studies. AI transfer learning research, using a microbiome-based random forest model, was conducted as a diagnosis tool, and the results demonstrated that the technique was able to detect a broad spectrum of cancer types using blood samples ( Xu et al, 2023 ). Huang et al (2023) developed an AI machine learning approach that examined colony morphology and genomic data through an open-source high-throughput robotic strain isolation platform for the rapid generation of cultured biobanks.…”

Machine learning: a powerful tool for identifying key microbial agents associated with specific cancer types

“…Deep learning (DL) belongs to a subclass of supervised learning, which has attracted increasing attention in microbiome research in recent years due to its unique advantages in processing large amounts of high and complex data. For microbiome data, the input features are relative abundances instead of images, which can be used to build DL models that classify the outcomes into presence or absence of disease state ( Xu et al, 2023 ). The most commonly used form of DL is artificial neural networks (ANN) ( Guo et al, 2020 ).…”

“…Coincidentally, Xu et al (2023) applied a large amount of microbial data from 21 cancer types, including 11,819 tissue samples (metagenomic data and 16S rRNA sequencing data) and 1,845 blood samples (metagenomic data), to construct DeepMicroCancer. The DeepMicroCancer is a set of tissue/blood microbiome based RF and tissue-blood microbiome based transfer learning models that can be used to diagnose a wide range of cancer types.…”

“…The top contributor in the blood RF model is Herbaspirillum , which has been reported to be isolated from multiple cancer patients ( Chemaly et al, 2015 ; Gungor et al, 2020 ; Suwantarat et al, 2015 ). In addition, the tissue–blood transfer learning (TL) model has a precision of 0.5 or higher for most cancers, and the multiclassification averaged AUC reached 0.89, which performed better than the RF model in predicting cancer types in small samples ( Xu et al, 2023 ). These findings suggest a new class of microbial-based cancer diagnostics, providing a unique opportunity to develop cancer diagnostics.…”