Human Gut Microbiome Biomarkers for Prediction of Depression

Abstract

Introduction

Although depression is the most common mental disorder in the world [1], the methods used for its detection and treatment are still limited (mainly based on clinical examination and subjective evaluation) [2]. This could be improved with non-invasive, quantitative tests based on biomarkers [2]. A potentially promising and relatively under-researched target for biomarker discovery in depression is the gut microbiome. Gut microbiome is one of the most important parts of the gut-brain axis (GBA), which represents the bidirectional pathways between the gut and the brain and appears to be dysfunctional in depression [1].

In our research we will acquire sequenced gut microbiomes and associated metadata of individuals with depression and healthy controls, pre-process them, identify potential biomarkers and use them to make a classification model for prediction of depression.

Methods

The acquired data is a subset of the Flemish Gut Flora Project dataset [3] and it contains faecal metagenomic data and metadata (age, sex, BMI, BSS, RAND) from 157 subjects (M = 50, SD = 12,96, 38% male), 80 of which have depression, 7 treatment-resistant depression and 70 are healthy controls. The sequencing data was pre-processed with bioBakery and mothur tools, which extract information about taxonomy, diversity, functional genes, enzyme reactions, metabolic pathways, and predicted metabolites from the sequenced microbiomes. We will try to identify which of these levels of information are significantly different between healthy individuals and those with depression and then use them to make a classification model in Orange, trying out different algorithms, settings, and optimality criteria to find the most accurate model.

Expected Results

We expect to identify biomarkers and create accurate classification models using various levels of microbiome information that will be able to distinguish between healthy individuals and those with depression. This will help shine light on the role of microbiome in the aetiology and pathogenesis of depression and highlight the importance and usefulness of implementing machine learning methods in microbiome and depression research.

References

[1] I. Łoniewski et al., “Major Depressive Disorder and gut microbiota – Association not causation. A scoping review”, Progress in Neuropsychopharmacology & Biological Psychiatry, vol. 106, 2021.

[2] Y. Hacimusalar and E. Eşel, “Suggested Biomarkers for Major Depressive Disorder”, Archives of Neuropsychiatry, vol. 55, no. 3, pp. 280-290, 2018.

[3] M. Valles-Colomer et al., "The neuroactive potential of the human gut microbiota in quality of life and depression", Nature Microbiology, vol. 4, pp. 623-632, 2019.