The Use of Machine Learning in Gut Microbiome Research A Scoping Review

This project explores the use of machine learning techniques in gut microbiome research, with a focus on Autism Spectrum Disorder (ASD). The study utilizes the "Human Gut Microbiome with ASD" dataset from Kaggle to investigate the relationship between gut microbiome features and ASD. The target variable for the classification is created using the Mann-Whitney U Test, where the dataset is divided into two groups: Group A (representing individuals with autism) and Group B (representing individuals without autism). The Mann-Whitney U Test is used to identify statistically significant differences in microbiome features between the two groups, which serves as the basis for the target labels. Machine learning models are then applied to classify the microbiome data, aiming to identify biomarkers associated with ASD.

This research highlights the potential of leveraging microbiome analysis to advance the understanding of ASD and its diagnosis. By training models on features extracted from microbiome data, the project seeks to enhance the classification accuracy for identifying ASD in individuals based on microbial compositions. The outcomes of this study may pave the way for more personalized approaches to autism detection and intervention. The project utilizes a Flask-based web interface to allow users to input data and obtain classification results. The findings aim to demonstrate the practical use of machine learning in microbiome research, offering valuable insights into the role of the microbiome in neurological conditions.

Keywords: Mann-Whitney U Test , machine learning, gut microbiome, Autism Spectrum Disorder, classification, biomarkers, microbiome data, ASD, machine learning models, Flask, web interface.