Artificial intelligence-based personalized diet: A pilot clinical study for irritable bowel syndrome.

We enrolled consecutive IBS-M patients (n = 25) according to Rome IV criteria. Fecal samples were obtained from all patients twice (pre-and post-intervention) and high-throughput 16S rRNA sequencing was performed. Six weeks of personalized nutrition diet (n = 14) for group 1 and a standard IBS diet (n = 11) for group 2 were followed. AI-based diet was designed based on optimizing a personalized nutritional strategy by an algorithm regarding individual gut microbiome features. The IBS-SSS evaluation for pre- and post-intervention exhibited significant improvement (p < .02 and p < .001 for the standard IBS diet and personalized nutrition groups, respectively). While the IBS-SSS evaluation changed to moderate from severe in 78% (11 out of 14) of the personalized nutrition group, no such change was observed in the standard IBS diet group. A statistically significant increase in the Faecalibacterium genus was observed in the personalized nutrition group (p = .04). Bacteroides and putatively probiotic genus Propionibacterium were increased in the personalized nutrition group. The change (delta) values in IBS-SSS scores (before-after) in personalized nutrition and standard IBS diet groups are significantly higher in the personalized nutrition group. AI-based personalized microbiome modulation through diet significantly improves IBS-related symptoms in patients with IBS-M. Further large-scale, randomized placebo-controlled trials with long-term follow-up (durability) are needed.

Discovering Potential Taxonomic Biomarkers of Type 2 Diabetes From Human Gut Microbiota via Different Feature Selection Methods.

Human gut microbiota is a complex community of organisms including trillions of bacteria. While these microorganisms are considered as essential regulators of our immune system, some of them can cause several diseases. In recent years, next-generation sequencing technologies accelerated the discovery of human gut microbiota. In this respect, the use of machine learning techniques became popular to analyze disease-associated metagenomics datasets. Type 2 diabetes (T2D) is a chronic disease and affects millions of people around the world. Since the early diagnosis in T2D is important for effective treatment, there is an utmost need to develop a classification technique that can accelerate T2D diagnosis. In this study, using T2D-associated metagenomics data, we aim to develop a classification model to facilitate T2D diagnosis and to discover T2D-associated biomarkers. The sequencing data of T2D patients and healthy individuals were taken from a metagenome-wide association study and categorized into disease states. The sequencing reads were assigned to taxa, and the identified species are used to train and test our model. To deal with the high dimensionality of features, we applied robust feature selection algorithms such as Conditional Mutual Information Maximization, Maximum Relevance and Minimum Redundancy, Correlation Based Feature Selection, and select K best approach. To test the performance of the classification based on the features that are selected by different methods, we used random forest classifier with 100-fold Monte Carlo cross-validation. In our experiments, we observed that 15 commonly selected features have a considerable effect in terms of minimizing the microbiota used for the diagnosis of T2D and thus reducing the time and cost. When we perform biological validation of these identified species, we found that some of them are known as related to T2D development mechanisms and we identified additional species as potential biomarkers. Additionally, we attempted to find the subgroups of T2D patients using k-means clustering. In summary, this study utilizes several supervised and unsupervised machine learning algorithms to increase the diagnostic accuracy of T2D, investigates potential biomarkers of T2D, and finds out which subset of microbiota is more informative than other taxa by applying state-of-the art feature selection methods.

Information Theoretic Metagenome Assembly Allows the Discovery of Disease Biomarkers in Human Microbiome.

Quantitative metagenomics is an important field that has delivered successful microbiome biomarkers associated with host phenotypes. The current convention mainly depends on unsupervised assembly of metagenomic contigs with a possibility of leaving interesting genetic material unassembled. Additionally, biomarkers are commonly defined on the differential relative abundance of compositional or functional units. Accumulating evidence supports that microbial genetic variations are as important as the differential abundance content, implying the need for novel methods accounting for the genetic variations in metagenomics studies. We propose an information theoretic metagenome assembly algorithm, discovering genomic fragments with maximal self-information, defined by the empirical distributions of nucleotides across the phenotypes and quantified with the help of statistical tests. Our algorithm infers fragments populating the most informative genetic variants in a single contig, named supervariant fragments. Experiments on simulated metagenomes, as well as on a colorectal cancer and an atherosclerotic cardiovascular disease dataset consistently discovered sequences strongly associated with the disease phenotypes. Moreover, the discriminatory power of these putative biomarkers was mainly attributed to the genetic variations rather than relative abundance. Our results support that a focus on metagenomics methods considering microbiome population genetics might be useful in discovering disease biomarkers with a great potential of translating to molecular diagnostics and biotherapeutics applications.

MIMOSA: Algorithms for Microbial Profiling.

A significant goal of the study of metagenomes obtained from an environment is to find the microbial diversity and the abundance of each organism in the community. Phylotyping and binning methods which address this problem generally operate using either marker sequences or by classifying each genome fragment individually. However, these approaches might not use all the information contained in the metagenome. We propose an approach based on a Multiple Input Multiple Output (MIMO) communication system model. Results from two different implementations of this approach, one using DNA-DNA hybridization simulations and one using short read mapping are evaluated using simulated and actual metagenomes and compared with other methods of phylotyping. The proposed approaches generally performed better under different scenarios including pathogen detection tasks of community complexity and low and high sequencing coverage while being highly computationally effective. The resulting framework can be integrated to metagenome analysis pipelines for phylogenetic diversity estimation. The approach is modular so that techniques other than hybridization simulations and short read mapping may be integrated. We have observed that even for low coverage samples, the method provides accurate estimates. Therefore, the use of the proposed strategy could enable the task of exploring biodiversity with limited resources.

Dynamic programming.

Independent scoring of the aligned sections to determine the quality of biological sequence alignments enables recursive definitions of the overall alignment score. This property is not only biologically meaningful but it also provides the opportunity to find the optimal alignments using dynamic programming-based algorithms. Dynamic programming is an efficient problem solving technique for a class of problems that can be solved by dividing into overlapping subproblems. Pairwise sequence alignment techniques such as Needleman-Wunsch and Smith-Waterman algorithms are applications of dynamic programming on pairwise sequence alignment problems. These algorithms offer polynomial time and space solutions. In this chapter, we introduce the basic dynamic programming solutions for global, semi-global, and local alignment problems. Algorithmic improvements offering quadratic-time and linear-space programs and approximate solutions with space-reduction and seeding heuristics are discussed. We finally introduce the application of these techniques on multiple sequence alignment briefly.