The number of microbiome-related studies has notably increased the availability of data on human microbiome composition and function. These studies provide the essential material to deeply explore host-microbiome associations and their relation to the development and progression of various complex diseases. Improved data-analytical tools are needed to exploit all information from these biological datasets, taking into account the peculiarities of microbiome data, i.e., compositional, heterogeneous and sparse nature of these datasets. The possibility of predicting host-phenotypes based on taxonomy-informed feature selection to establish an association between microbiome and predict disease states is beneficial for personalized medicine. In this regard, machine learning (ML) provides new insights into the development of models that can be used to predict outputs, such as classification and prediction in microbiology, infer host phenotypes to predict diseases and use microbial communities to stratify patients by their characterization of state-specific microbial signatures. Here we review the state-of-the-art ML methods and respective software applied in human microbiome studies, performed as part of the COST Action ML4Microbiome activities. This scoping review focuses on the application of ML in microbiome studies related to association and clinical use for diagnostics, prognostics, and therapeutics. Although the data presented here is more related to the bacterial community, many algorithms could be applied in general, regardless of the feature type. This literature and software review covering this broad topic is aligned with the scoping review methodology. The manual identification of data sources has been complemented with: (1) automated publication search through digital libraries of the three major publishers using natural language processing (NLP) Toolkit, and (2) an automated identification of relevant software repositories on GitHub and ranking of the related research papers relying on learning to rank approach.

The increasing prevalence of diet-related diseases calls for an improvement in nutritional advice. Personalized nutrition aims to solve this problem by adapting dietary and lifestyle guidelines to the unique circumstances of each individual. With the latest advances in technology and data science, researchers can now automatically collect and analyze large amounts of data from a variety of sources, including wearable and smart devices. By combining these diverse data, more comprehensive insights of the human body and its diseases can be achieved. However, there are still major challenges to overcome, including the need for more robust data and standardization of methodologies for better subject monitoring and assessment. Here, we present the AI4Food database (AI4FoodDB), which gathers data from a nutritional weight loss intervention monitoring 100 overweight and obese participants during 1 month. Data acquisition involved manual traditional approaches, novel digital methods and the collection of biological samples, obtaining: (i) biological samples at the beginning and the end of the intervention, (ii) anthropometric measurements every 2 weeks, (iii) lifestyle and nutritional questionnaires at two different time points and (iv) continuous digital measurements for 2 weeks. To the best of our knowledge, AI4FoodDB is the first public database that centralizes food images, wearable sensors, validated questionnaires and biological samples from the same intervention. AI4FoodDB thus has immense potential for fostering the advancement of automatic and novel artificial intelligence techniques in the field of personalized care. Moreover, the collected information will yield valuable insights into the relationships between different variables and health outcomes, allowing researchers to generate and test new hypotheses, identify novel biomarkers and digital endpoints, and explore how different lifestyle, biological and digital factors impact health. The aim of this article is to describe the datasets included in AI4FoodDB and to outline the potential that they hold for precision health research. Database URL https://github.com/AI4Food/AI4FoodDB

The objectives of our study were to describe the characteristics of patients with Candida guilliermondii candidemia and to perform an in-depth microbiological characterization of isolates and compare them with those of patients with C. albicans candidemia. We described the risk factors and outcomes of 22 patients with candidemia caused by the C. guilliermondii complex. Incident isolates were identified using molecular techniques, and susceptibility to fluconazole, anidulafungin, and micafungin was studied. Biofilm formation was measured using the crystal violet assay (biomass production) and the XTT reduction assay (metabolic activity), and virulence was studied using the Galleria mellonella model. Biofilm formation was compared with that observed for C. albicans . The main conditions predisposing to infection were malignancy (68%), immunosuppressive therapy (59%), and neutropenia (18%). Clinical presentation of candidemia was less severe in patients infected by the C. guilliermondii complex than in patients infected by C. albicans , and 30-day mortality was lower in C. guilliermondii patients (13.6% versus 33.9%, respectively; P = 0.049). Isolates were identified as C. guilliermondii sensu stricto ( n = 17) and Candida fermentati ( n = 5). The isolates produced biofilms with low metabolic activity and moderate biomass. The G. mellonella model showed that C. guilliermondii was less virulent than C. albicans (mean of 6 days versus 1 day of survival, respectively; P < 0.001). Patients with candidemia caused by the C. guilliermondii complex had severe and debilitating underlying conditions. Overall, the isolates showed diminished susceptibility to fluconazole and echinocandins, although poor biofilm formation and the low virulence were associated with a favorable outcome.

Accurate knowledge of fungemia epidemiology requires identification of strains to the molecular level. Various studies have shown that the rate of resistance to fluconazole ranges from 2.5% to 9% in Candida spp. isolated from blood samples. However, trends in antifungal resistance have received little attention and have been studied only using CLSI M27-A3 methodology. We assessed the fungemia epidemiology in a large tertiary care institution in Madrid, Spain, by identifying isolates to the molecular level and performing antifungal susceptibility testing according to the updated breakpoints of European Committee for Antimicrobial Susceptibility Testing (EUCAST) definitive document (EDef) 7.2. We studied 613 isolates causing 598 episodes of fungemia in 544 patients admitted to our hospital (January 2007 to December 2013). Strains were identified after amplification and sequencing of the ITS1-5.8S-ITS2 region and further tested for in vitro susceptibility to amphotericin B, fluconazole, posaconazole, voriconazole, micafungin, and anidulafungin. Resistance was defined using EUCAST species-specific breakpoints, and epidemiological cutoff values (ECOFFs) were applied as tentative breakpoints. Most episodes were caused by Candida albicans (46%), Candida parapsilosis (28.7%), Candida glabrata (9.8%), and Candida tropicalis (8%). Molecular identification enabled us to better detect cryptic species of Candida guilliermondii and C. parapsilosis complexes and episodes of polyfungal fungemia. The overall percentage of fluconazole-resistant isolates was 5%, although it was higher in C. glabrata (8.6%) and non-Candida yeast isolates (47.4%). The rate of resistance to echinocandins was 4.4% and was mainly due to the presence of intrinsically resistant non-Candida species. Resistance mainly affected non-Candida yeasts. The rate of resistance to fluconazole and echinocandins did not change considerably during the study period.