Biology subjects

Aghdam, R.

Publications and source records attributed to Aghdam, R..

8 recordsLinked to original sources

bioRxiv2026

PhyMapNet: A Phylogeny-Guided Bayesian Framework for Reliable Microbiome Network Inference

Understanding microbial interactions is essential for revealing the ecological structure and functional organization of microbiomes. However, network inference remains challenging due to the compositional, sparse, and high-dimensional nature of microbiome data, as well as the lack of gold-standard interaction benchmarks. Here, we introduce PhyMapNet, a Bayesian Gaussian Graphical Model framework that explicitly integrates phylogenetic information to infer microbiome networks. We introduce PhyMapNet, a phylogeny-aware method for reconstructing microbial association networks. PhyMapNet infers conditional dependencies among taxa by estimating a sparse precision matrix, while incorporating evolutionary relatedness through a kernel defined on phylogenetic distances. This integration of biological prior information improves the robustness and interpretability of the inferred networks. We evaluated PhyMapNet on two real-world microbiome datasets (Smoking and Caffeine), performing extensive robustness analyses under bootstrap and noisy perturbations, which demonstrated the reproducibility of the algorithm. The results also revealed that network structure can vary with parameter selection, highlighting the algorithms sensitivity to hyperparameters. Because PhyMapNet is computationally efficient and can explore thousands of parameter configurations within an hour, we developed a tuning-free framework that constructs reliable consensus networks by aggregating results across a broad hyperparameter space. This strategy yields consensus microbiome networks with controllable sparsity, offering a practical way to adjust network density while retaining highly stable edges. We then compared these consensus networks with those inferred by nine established methods and observed statistically significant overlap. To support reproducibility and practical use, we provide an open-source R package implementation of PhyMapNet.

bioinformatics↗

bioRxiv2025

Evaluating microbial network inference methods: Moving beyond synthetic data with reproducibility-driven benchmarks

Microbial network inference is an essential approach for revealing complex interactions within microbial communities. However, the lack of experimentally validated gold standards presents a significant obstacle in evaluating the accuracy and biological relevance of inferred networks. This study delivers a comprehensive comparative assessment of six widely used microbial network inference algorithms--gCoda, OIPCQ, S-E(glasso), S-E(mb), SPRING, and SparCC--using four diverse real-world microbiome datasets alongside multiple types of generated data, including synthetic, noisy, and bootstrap-derived datasets. Our evaluation framework extends beyond conventional synthetic benchmarking by emphasizing reproducibility-focused assessments grounded in biologically realistic perturbations. We show that bootstrap resampling and low-level noisy datasets ([≤]10% perturbation) effectively preserve key statistical properties of real microbiome data, such as diversity indices, abundance distributions, and sparsity patterns. Conversely, synthetic datasets generated via the widely used SPIEC-EASI method exhibit substantial divergence from real data across these metrics. We find that while SparCC demonstrates superior robustness across varying data conditions, other methods tend to produce inflated performance metrics when evaluated on unrealistic synthetic networks. Notably, several algorithms fail to distinguish between structured and random networks, highlighting issues of structural insensitivity and the limitations of overreliance on synthetic benchmarks. To address these challenges, we propose a reproducibility-centered benchmarking framework that prioritizes real-data-derived perturbations and mandates rigorous statistical validation of synthetic datasets before their use. This work provides critical insights and practical guidance for the microbiome research community, aiming to foster more reliable and ecologically meaningful microbial network inference in the absence of a true ground truth.

bioinformatics↗

bioRxiv2025

CMiNet: An R Package and User-Friendly Shiny App for Constructing Consensus Microbiome Networks

O_LIMicrobial networks offer critical insights into community structure, ecological interactions, and host-microbe dynamics. However, constructing reliable microbiome networks remains challenging due to variability among existing inference methods, limited overlap between networks, and the absence of a gold standard for validation. C_LIO_LIWe developed CMiNet (https://cminet.wid.wisc.edu), an interactive Shiny application and R package that enables consensus microbiome network construction by integrating up to ten widely used inference algorithms. CMiNet supports both correlation-based and conditional dependence-based methods and provides users with flexible options to construct individual or consensus networks across different approaches. C_LIO_LICMiNet employs a consensus-based filtering strategy that retains only edges supported by multiple methods, leading to more robust network structures that better reflect underlying biological interactions. This approach enhances reproducibility, minimizes method-specific biases, and improves biological interpretability. Its user-friendly interface allows researchers to upload data, customize network construction parameters, visualize networks interactively, and export results--without requiring programming expertise. C_LIO_LIWe demonstrated CMiNet using both gut and soil microbiome datasets. In the soil microbiome application, robust network construction enabled the consistent identification of disease-associated taxa. Specifically, we identified a set of key taxa, including Ktedonobacteria, Acidobacteriae, Vicinamibacteria, MB-A2-108, Planctomycetes, and Anaerolineae, selected by multiple independent methods. C_LI

systems biology↗

bioRxiv2025

SimMapNet: A Bayesian Framework for Gene Regulatory Network Inference Using Gene Ontology Similarities as External Hint

MotivationGene regulatory network (GRN) reconstruction is a fundamental challenge in computational biology, and is crucial for understanding gene interactions. In this study, we aim to incorporate Gene Ontology (GO) similarities into the construction of GRNs. Our key assumption is that genes with higher similarity in Molecular Function, Biological Process, or Cellular Component categories are more likely to be functionally related and, therefore, more likely to be connected in the network. We introduce SimMapNet, a Bayesian framework that estimates the precision matrix, which serves as the adjacency matrix in a Gaussian graphical model (GGM) for GRN inference. SimMapNet enhances network inference by integrating GO similarities, which inform the hyperparameters of the prior distribution through a kernel function, incorporating biological prior knowledge in a principled manner. ResultsWe evaluate SimMapNet on three datasets: two datasets from the SOS DNA-repair response pathway in Escherichia coli and one dataset from Drosophila melanogaster. The results demonstrate the algorithms superior performance compared to state-of-the-art methods such as GLASSO, GENIE3, and KBOOST in terms of F1-score. SimMapNet has low time complexity, making it suitable for constructing large networks. Our simulation results confirm that SimMapNet is particularly well-suited for scenarios with limited sample sizes, where traditional methods often struggle. Availability and implementationThe datasets and R package of SimMapNet are available in the github repository, https://github.com/maryam-shahdoust/SimMapNet.

bioinformatics↗

bioRxiv2025

Leveraging Bayesian Networks for Consensus Network Construction and Multi-Method Feature Selection to Decode Disease Prediction

Constructing reliable microbiome co-occurrence networks and identifying disease-associated taxa remain major challenges in microbiome research due to variability introduced by different inference algorithms. To overcome these limitations, we present CMIMN, a novel R package that uses a Bayesian network framework based on conditional mutual information to infer robust microbial interaction networks. To further enhance reliability, we construct a consensus microbiome network by integrating results from CMIMN and three widely used methods--SPIEC-EASI, SPRING, and SPARCC. This consensus approach, which overlays and weights edges shared across methods, reduces inconsistencies and provides a more biologically meaningful view of microbial relationships. In addition, we introduce a multi-method framework for identifying disease-associated microbial taxa by combining machine learning and network-based feature selection. Our ML pipeline applies distinct algorithms and identifies key taxa based on their consistent importance across models. Complementing this, we employ two network-based strategies that prioritize taxa based on centrality differences between clean tubers and scab-infected tubers networks, as well as a composite scoring system that ranks nodes using integrated network metrics. Our results show that CMIMN achieves high robustness in network inference, and that the consensus network further improves stability and interpretability. The multi-method feature selection framework enhances confidence in identifying biologically relevant taxa linked to potato common scab disease. Notably, we identify Bacteroidota, WPS-2, and Proteobacteria at the Phylum level; Actinobacteria, AD3, Bacilli, Anaerolineae, and Ktedonobacteria at the Class level; and C0119, Defluviicoccales, Bacteroidales, and Ktedonobacterales at the Order level as key taxa associated with disease status.

systems biology↗

bioRxiv2024

Unraveling Keystone Taxa: Interactions Within Microbial Networks and Environmental Dynamics in Lake Mendota

Microbial communities in freshwater ecosystems drive critical biogeochemical cycles, nutrient transformations, and energy flows essential for ecosystem stability. Yet, in the face of accelerating environmental changes, the responses of these microbial networks to spatial and temporal shifts remain underexplored, particularly with rising anoxia. We investigated the microbial ecosystems of Lake Mendota, Wisconsin, USA, through comprehensive metagenomic and metatranscriptomic analyses to elucidate their adaptations to environmental fluctuations across temporal and spatial dimensions. Employing tools like Sparse Inverse Covariance Estimation for Ecological Association Inference (SPIEC-EASI) and Conditional Auto-Regressive Least Absolute Shrinkage and Selection Operator (CARlasso), we identified key microbial taxa and their interactions with environmental parameters such as depth, temperature, pH, and dissolved oxygen. Our findings reveal that biological interactions more than environmental variables shape microbial community assembly and function. Specifically, keystone taxa from the phylum Bacteroidota emerged as pivotal in nutrient cycling and organic matter decomposition, processes crucial for sustaining water quality. Notably, these keystone taxa demonstrate dynamic adaptability, suggesting that microbial networks can rapidly adjust to changes in composition, a trait essential for resilience in the face of warming temperatures and altered precipitation patterns. This study provides critical insights into the resilience and adaptability of freshwater microbiomes, highlighting the role of microbial interactions in maintaining ecosystem health. By understanding how these microbial networks respond to environmental pressures, we can better predict shifts in microbial dynamics and anticipate the broader ecological impacts of climate change on freshwater systems. ImportanceThis research underscores the critical role of keystone taxa in freshwater ecosystems, highlighting how these organisms maintain water quality and contribute to the stability of aquatic environments. Understanding the ecological roles of these taxa is essential for developing strategies to manage ecosystems and conserve freshwater resources, particularly in the face of ongoing environmental challenges like climate change. The insights provided by this study not only enhance our comprehension of microbial interactions but also support effective ecosystem management and conservation efforts.

microbiology↗

bioRxiv2024

HighDimMixedModels.jl: Robust High Dimensional Mixed Models across Omics Data

High dimensional mixed-effect models are an increasingly important form of regression in modern biology, in which the number of variables often matches or exceeds the number of samples, which are collected in groups or clusters. The penalized likelihood approach to fitting these models relies on a coordinate gradient descent (CGD) algorithm that lacks guarantees of convergence to a global optimum. Here, we study empirically the behavior of the algorithm across a number of common study types in modern omics datatypes. In particular, we study the empirical performance of high dimensional mixed-effect models fit to data simulated to mimic the features of transcriptome, genome-wide association, and microbiome data. In addition, we study the performance of the model on real data from each of these study types. To facilitate these simulations, we implement the algorithm in an open source Julia package HighDimMixedModels.jl. We compare the performance of two commonly used penalties, namely LASSO and SCAD, within the HighDimMixedModels.jl framework. Our results demonstrate that the SCAD penalty consistently outperforms LASSO in terms of both variable selection and estimation accuracy across omics data. Through our comprehensive analysis, we illuminate the intricate relationship between algorithmic behavior, penalty selection, and dataset properties such as the correlation structure among features, providing valuable insights for researchers employing high dimensional mixed-effect models in biological investigations. Author SummaryHigh dimensional mixed-effect models are increasingly indispensable in modern biology, particularly in omics studies, where the number of variables often equals or surpasses the number of samples, and data are collected in clusters or groups. In our research, we concentrate on the penalized likelihood approach to fitting these models, employing a coordinate gradient descent (CGD) algorithm. While CGD is a widely used optimization technique, its convergence to a global optimum lack guarantees, prompting our empirical investigation into its behavior across various study types common in modern omics datasets. Our study provides insights into the performance of high dimensional mixed-effect models fitted to data simulating transcriptome, genome-wide association, and microbiome datasets. Additionally, we evaluate the models performance on real datasets from each of these study types. To facilitate reproducibility and further research, we have implemented the algorithm in an open-source Julia package, HighDimMixedModels.jl. Notably, HighDimMixedModels.jl stands out as the first package capable of seamlessly handling various omics datasets without errors, offering a user-friendly solution for researchers across disciplines. While numerous software packages are available for implementing high dimensional mixed-effects models on omics data, there is currently no comprehensive review source summarizing all methods. We provide a table summarizing existing methods, available in the Supplementary Material.

systems biology↗

bioRxiv2021

NN-RNALoc: neural network-based model for prediction of mRNA sub-cellular localization using distance-based sub-sequence profiles

Localization of messenger RNAs (mRNA) as a widely observed phenomenon is considered as an efficient way to target proteins to a specific region of a cell and is also known as a strategy for gene regulation. The importance of correct intracellular RNA placement in the development of embryonic and neural dendrites has long been demonstrated in former studies. Improper localization of RNA in the cell, which has been shown to occur due to a variety of reasons, including mutations in trans-regulatory elements, is also associated with the occurrence of some neuromuscular diseases as well as cancer. We propose NN-RNALoc, a neural network-based model to predict the cellular location of mRNAs. The features extracted from mRNA sequences along with the information gathered from their proteins are fed to this prediction model. We introduce a novel distance-based sub-sequence profile for representation of RNA sequences which is more memory and time efficient and comparying to the k-mer frequencies, can possibly better encode sequences when the distance k increases. The performance of NN-RNALoc on the following benchmark datsets CeFra-seq and RNALocate, is compared to the results achieved by two powerful prediction models that were proposed in former studies named as mRNALoc and RNATracker The results reveal that the employment of protein-protein interaction information, which plays a crucial role in many biological functions, together with the novel distance-based sub-sequence profiles of mRNA sequences, leads to a more accurate prediction model. Besides, NN-RNALoc significantly reduces the required computing time compared to previous studies. Source code and data used in this study are available at: https://github.com/NeginBabaiha/NN-RNALoc

bioinformatics↗