Microbiome-based disease prediction with multimodal variational information bottlenecks

Abstract

Scientific research is shedding light on the interaction of the gut microbiome with the human host and on its role in human health. Existing machine learning methods have shown great potential in discriminating healthy from diseased microbiome states. Most of them leverage shotgun metagenomic sequencing to extract gut microbial species-relative abundances or strain-level markers. Each of these gut microbial profiling modalities showed diagnostic potential when tested separately; however, no existing approach combines them in a single predictive framework. Here, we propose the Multimodal Variational Information Bottleneck (MVIB), a novel deep learning model capable of learning a joint representation of multiple heterogeneous data modalities. MVIB achieves competitive classification performance while being faster than existing methods. Additionally, MVIB offers interpretable results. Our model adopts an information theoretic interpretation of deep neural networks and computes a joint stochastic encoding of different input data modalities. We use MVIB to predict whether human hosts are affected by a certain disease by jointly analysing gut microbial species-relative abundances and strain-level markers. MVIB is evaluated on human gut metagenomic samples from 11 publicly available disease cohorts covering 6 different diseases. We achieve high performance (0.80 < ROC AUC < 0.95) on 5 cohorts and at least medium performance on the remaining ones. We adopt a saliency technique to interpret the output of MVIB and identify the most relevant microbial species and strain-level markers to the model’s predictions. We also perform cross-study generalisation experiments, where we train and test MVIB on different cohorts of the same disease, and overall we achieve comparable results to the baseline approach, i.e. the Random Forest. Further, we evaluate our model by adding metabolomic data derived from mass spectrometry as a third input modality. Our method is scalable with respect to input data modalities and has an average training time of < 1.4 seconds. The source code and the datasets used in this work are publicly available.

Attentive Variational Information Bottleneck for TCR–peptide interaction prediction

“…Similar to previous works ( Grazioli et al , 2022a ; Wu and Goodman, 2018 ), the single-sequence posteriors are modeled as Gaussian distributions with diagonal structure: . By stacking the parameters (represented as column-vectors) and of the latent prior with the and for all available sequences, we define the following two matrices and , where d Z is the dimensionality of the latent single-sequence posteriors: …”

“…This allows using stochastic gradient descent to optimize the objective. Various multimodal generalizations of the VIB have been recently proposed: the Multimodal Variational Information Bottleneck (MVIB) ( Grazioli et al , 2022a ) and DeepIMV ( Lee and Schaar, 2021 ). Both MVIB and DeepIMV adopt the PoE to estimate a joint multimodal latent encoding distribution from the unimodal latent encoding distributions.…”

Machine learning and deep learning applications in microbiome research

“…As presented by Reiman and Dai [61], a bimodal autoencoder can integrate diet and microbial composition to predict the microbial dynamics response to dietary change. Grazioli et al [62] introduce a disease prediction model that relies on the product-of-experts approach to integrate the information from two autoencoders, each expert on a different modality: abundance (species-level) and presence (strain-level) features, respectively.…”

A Mini Review of Node Centrality Metrics in Biological Networks

“…Each of the nodes within the shortest path trees is identified as "bottlenecks" [96,97]. There exists a relationship between bottlenecks (high centrality nodes) and essentiality [98][99][100]. It is crucial to note that the same node can function as a bottleneck for multiple shortest path trees.…”

“…It has been found that betweenness centrality may accurately identify network hub nodes that can enhance the transmission efficiency of data [104]. The betweenness of a protein reveals the protein's potential to facilitate communication among a variety of proteins in the protein networks [98]. Proteins have high betweenness centralities are referred to as key connector proteins with crucial functional and dynamic features [105,106], such as metabolites that regulate the flux between two large metabolic modules.…”