mbalances in the gut microbiome have been linked to conditions such as inflammatory bowel disease, diabetes, and cancer. While metagenomics and amplicon sequencing are commonly used to study the microbiome, they do not capture all layers of microbial functions. Other meta-omics data can provide more insights, but these are more costly and laborious to procure. The growing availability of paired meta-omics data offers an opportunity to develop machine learning models that can infer connections between metagenomics data and other forms of meta-omics data, enabling the prediction of these other forms of meta-omics data from metagenomics. We evaluated several machine learning models for predicting meta-omics features from various meta-omics inputs. Simpler architectures such as elastic net regression and random forests generated reliable predictions of transcript and metabolite abundances, with correlations of up to 0.77 and 0.74, respectively, but predicting protein profiles was more challenging. We also identified a core set of well-predicted features for each meta-omics output type, and showed that multi-output regression neural networks performed similarly when trained using fewer output features. Lastly, our experiments demonstrated that predicted features can be used for the downstream task of inflammatory bowel disease classification, with performance comparable to that of experimental data.

Bacterial resistance to antimicrobials is a global health threat. Within the One Health context, water from regions with high antibiotic usage, such as clinical and urban areas, collects at wastewater treatment plants (WWTPs). In the WWTP, the activated sludge becomes a complex environment where various antimicrobials and microorganisms converge. While significant research has focused on the influent, activated sludge, and effluent, upstream and downstream sectors around the WWTP are often neglected. We conducted a systematic analysis using five publicly available metagenomic datasets (n=164) from different WWTP sectors and adjacent freshwater systems: upstream (n=14), influent (n=14), activated sludge (n=109), effluent (n=14), and downstream (n=13) to identify and characterise the microbiome, resistome, and mobilome. Opportunistic pathogenic bacteria, such as Pseudomonas, Aeromonas, and Acidovorax, were found in all WWTP sectors, with abundances exceeding 9% in the influent. ESKAPE pathogens, including Klebsiella pneumoniae and Enterobacter species, were identified in the effluent with abundances over 1%. We detected 230 antibiotic resistance genes (ARGs) throughout the WWTP. FTU and CKO β-lactamase gene families dominated the upstream, effluent, and downstream sectors, while the OXA β-lactamase gene family was highly abundant in the influent and activated sludge. ARGs, such as the OXA β-lactamase gene family, were linked to plasmids. Class-1 integrons, associated with the sul1 gene, a marker for anthropogenic pollution, were prevalent in the effluent and downstream sectors. Integrative elements (ICEclc, Tn4371, and PGI2), linked to ARGs, were identified in all sectors, increasing AMR dissemination. These integrative elements conferred resistance to antibiotics, including sulfonamides, tetracyclines and carbapenems. Our findings highlight the presence of ARGs and mobile genetic elements in WWTPs and nearby freshwater systems, raising concerns about AMR transmission to humans, animals, and the environment. This study emphasises the need for effective AMR monitoring and strategies in wastewater treatment to protect public and environmental health.