Towards a metagenomics machine learning interpretable model for understanding the transition from adenoma to colorectal cancer

A review of machine learning methods for cancer characterization from microbiome data

Recent studies have shown that the microbiome can impact cancer development, progression, and response to therapies suggesting microbiome-based approaches for cancer characterization. As cancer-related signatures are complex and implicate many taxa, their discovery often requires Machine Learning approaches. This review discusses Machine Learning methods for cancer characterization from microbiome data. It focuses on the implications of choices undertaken during sample collection, feature selection and pre-processing. It also discusses ML model selection, guiding how to choose an ML model, and model validation. Finally, it enumerates current limitations and how these may be surpassed. Proposed methods, often based on Random Forests, show promising results, however insufficient for widespread clinical usage. Studies often report conflicting results mainly due to ML models with poor generalizability. We expect that evaluating models with expanded, hold-out datasets, removing technical artifacts, exploring representations of the microbiome other than taxonomical profiles, leveraging advances in deep learning, and developing ML models better adapted to the characteristics of microbiome data will improve the performance and generalizability of models and enable their usage in the clinic.

Deep learning methods in metagenomics: a review

The ever-decreasing cost of sequencing and the growing potential applications of metagenomics have led to an unprecedented surge in data generation. One of the most prevalent applications of metagenomics is the study of microbial environments, such as the human gut. The gut microbiome plays a crucial role in human health, providing vital information for patient diagnosis and prognosis. However, analysing metagenomic data remains challenging due to several factors, including reference catalogues, sparsity and compositionality. Deep learning (DL) enables novel and promising approaches that complement state-of-the-art microbiome pipelines. DL-based methods can address almost all aspects of microbiome analysis, including novel pathogen detection, sequence classification, patient stratification and disease prediction. Beyond generating predictive models, a key aspect of these methods is also their interpretability. This article reviews DL approaches in metagenomics, including convolutional networks, autoencoders and attention-based models. These methods aggregate contextualized data and pave the way for improved patient care and a better understanding of the microbiome's key role in our health.

A comprehensive overview of microbiome data in the light of machine learning applications: categorization, accessibility, and future directions

Metagenomics, Metabolomics, and Metaproteomics have significantly advanced our knowledge of microbial communities by providing culture-independent insights into their composition and functional potential. However, a critical challenge in this field is the lack of standard and comprehensive metadata associated with raw data, hindering the ability to perform robust data stratifications and consider confounding factors. In this comprehensive review, we categorize publicly available microbiome data into five types: shotgun sequencing, amplicon sequencing, metatranscriptomic, metabolomic, and metaproteomic data. We explore the importance of metadata for data reuse and address the challenges in collecting standardized metadata. We also, assess the limitations in metadata collection of existing public repositories collecting metagenomic data. This review emphasizes the vital role of metadata in interpreting and comparing datasets and highlights the need for standardized metadata protocols to fully leverage metagenomic data's potential. Furthermore, we explore future directions of implementation of Machine Learning (ML) in metadata retrieval, offering promising avenues for a deeper understanding of microbial communities and their ecological roles. Leveraging these tools will enhance our insights into microbial functional capabilities and ecological dynamics in diverse ecosystems. Finally, we emphasize the crucial metadata role in ML models development.

Integrating clinical and cross-cohort metagenomic features: a stable and non-invasive colorectal cancer and adenoma diagnostic model

Background: Dysbiosis is associated with colorectal cancer (CRC) and adenomas (CRA). However, the robustness of diagnostic models based on microbial signatures in multiple cohorts remains unsatisfactory. Materials and Methods: In this study, we used machine learning models to screen metagenomic signatures from the respective cross-cohort datasets of CRC and CRA (selected from CuratedMetagenomicData, each disease included 4 datasets). Then select a CRC and CRA data set from the CuratedMetagenomicData database and meet the requirements of having both metagenomic data and clinical data. This data set will be used to verify the inference that integrating clinical features can improve the performance of microbial disease prediction models. Results: After repeated verification, we selected 20 metagenomic features that performed well and were stably expressed within cross-cohorts to represent the diagnostic role of bacterial communities in CRC/CRA. The performance of the selected cross-cohort metagenomic features was stable for multi-regional and multi-ethnic populations (CRC, AUC: 0.817-0.867; CRA, AUC: 0.766-0.833). After clinical feature combination, AUC of our integrated CRC diagnostic model reached 0.939 (95% CI: 0.932-0.947, NRI=30%), and that of the CRA integrated model reached 0.925 (95%CI: 0.917-0.935, NRI=18%). Conclusion: In conclusion, the integrated model performed significantly better than single microbiome or clinical feature models in all cohorts. Integrating cross-cohort common discriminative microbial features with clinical features could help construct stable diagnostic models for early non-invasive screening for CRC and CRA.

Microbiomes, Their Function, and Cancer: How Metatranscriptomics Can Close the Knowledge Gap

The interaction between the microbial communities in the human body and the onset and progression of cancer has not been investigated until recently. The vast majority of the metagenomics research in this area has concentrated on the composition of microbiomes, attempting to link the overabundance or depletion of certain microorganisms to cancer proliferation, metastatic behaviour, and its resistance to therapies. However, studies elucidating the functional implications of the microbiome activity in cancer patients are still scarce; in particular, there is an overwhelming lack of studies assessing such implications directly, through analysis of the transcriptome of the bacterial community. This review summarises the contributions of metagenomics and metatranscriptomics to the knowledge of the microbial environment associated with several cancers; most importantly, it highlights all the advantages that metatranscriptomics has over metagenomics and suggests how such an approach can be leveraged to advance the knowledge of the cancer bacterial environment.

Progress in the diagnosis of lymph node metastasis in rectal cancer: a review

Historically, the chief focus of lymph node metastasis research has been molecular and clinical studies of a few essential pathways and genes. Recent years have seen a rapid accumulation of massive omics and imaging data catalyzed by the rapid development of advanced technologies. This rapid increase in data has driven improvements in the accuracy of diagnosis of lymph node metastasis, and its analysis further demands new methods and the opportunity to provide novel insights for basic research. In fact, the combination of omics data, imaging data, clinical medicine, and diagnostic methods has led to notable advances in our basic understanding and transformation of lymph node metastases in rectal cancer. Higher levels of integration will require a concerted effort among data scientists and clinicians. Herein, we review the current state and future challenges to advance the diagnosis of lymph node metastases in rectal cancer.

Multi-omics insights into the interplay between gut microbiota and colorectal cancer in the "microworld" age

Colorectal cancer (CRC) is a multifactorial heterogeneous disease largely due to both genetic predisposition and environmental factors including the gut microbiota, a dynamic microbial ecosystem inhabiting the gastrointestinal tract. Elucidation of the molecular mechanisms by which the gut microbiota interacts with the host may contribute to the pathogenesis, diagnosis, and promotion of CRC. However, deciphering the influence of genetic variants and interactions with the gut microbial ecosystem is rather challenging. Despite recent advancements in single omics analysis, the application of multi-omics approaches to integrate multiple layers of information in the microbiome and host to introduce effective prevention, diagnosis, and treatment strategies is still in its infancy. Here, we integrate host- and microbe-based multi-omics studies, respectively, to provide a strategy to explore potential causal relationships between gut microbiota and colorectal cancer. Specifically, we summarize the recent multi-omics studies such as metagenomics combined with metabolomics and metagenomics combined with genomics. Meanwhile, the sample size and sample types commonly used in multi-omics research, as well as the methods of data analysis, were also generalized. We highlight multiple layers of information from multi-omics that need to be verified by different types of models. Together, this review provides new insights into the clinical diagnosis and treatment of colorectal cancer patients.

Antitumor effects of fecal microbiota transplantation: Implications for microbiome modulation in cancer treatment

Fecal microbiome transplantation (FMT) from healthy donors is one of the techniques for restoration of the dysbiotic gut, which is increasingly being used to treat various diseases. Notably, mounting evidence in recent years revealed that FMT has made a breakthrough in the oncology treatment area, especially by improving immunotherapy efficacy to achieve antitumor effects. However, the mechanism of FMT in enhancing antitumor effects of immune checkpoint blockers (ICBs) has not yet been fully elucidated. This review systematically summarizes the role of microbes and their metabolites in the regulation of tumor immunity. We highlight the mechanism of action of FMT in the treatment of refractory tumors as well as in improving the efficacy of immunotherapy. Furthermore, we summarize ongoing clinical trials combining FMT with immunotherapy and further focus on refined protocols for the practice of FMT in cancer treatment, which could guide future directions and priorities of FMT scientific development.

Meta-Analysis and Validation of a Colorectal Cancer Risk Prediction Model Using Deep Sequenced Fecal Metagenomes

The gut microbiome is a potential modifiable risk factor for colorectal cancer (CRC). We re-analyzed all eight previously published stool sequencing data and conducted an MWAS meta-analysis. We used cross-validated LASSO predictive models to identify a microbiome signature for predicting the risk of CRC and precancerous lesions. These models were validated in a new study, Colorectal Cancer Screening (COLSCREEN), including 156 participants that were recruited in a CRC screening context. The MWAS meta-analysis identified 95 bacterial species that were statistically significantly associated with CRC (FDR < 0.05). The LASSO CRC predictive model obtained an area under the receiver operating characteristic curve (aROC) of 0.81 (95%CI: 0.78−0.83) and the validation in the COLSCREEN dataset was 0.75 (95%CI: 0.66−0.84). This model selected a total of 32 species. The aROC of this CRC-trained model to predict precancerous lesions was 0.52 (95%CI: 0.41−0.63). We have identified a signature of 32 bacterial species that have a good predictive accuracy to identify CRC but not precancerous lesions, suggesting that the identified microbes that were enriched or depleted in CRC are merely a consequence of the tumor. Further studies should focus on CRC as well as precancerous lesions with the intent to implement a microbiome signature in CRC screening programs.

Combination of Whole Genome Sequencing and Metagenomics for Microbiological Diagnostics

Whole genome sequencing (WGS) provides the highest resolution for genome-based species identification and can provide insight into the antimicrobial resistance and virulence potential of a single microbiological isolate during the diagnostic process. In contrast, metagenomic sequencing allows the analysis of DNA segments from multiple microorganisms within a community, either using an amplicon- or shotgun-based approach. However, WGS and shotgun metagenomic data are rarely combined, although such an approach may generate additive or synergistic information, critical for, e.g., patient management, infection control, and pathogen surveillance. To produce a combined workflow with actionable outputs, we need to understand the pre-to-post analytical process of both technologies. This will require specific databases storing interlinked sequencing and metadata, and also involves customized bioinformatic analytical pipelines. This review article will provide an overview of the critical steps and potential clinical application of combining WGS and metagenomics together for microbiological diagnosis.