skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: Latent Model-Based Clustering for Biological Discovery
LOVE, a robust, scalable latent model-based clustering method for biological discovery, can be used across a range of datasets to generate both overlapping and non-overlapping clusters. In our formulation, a cluster comprises variables associated with the same latent factor and is determined from an allocation matrix that indexes our latent model. We prove that the allocation matrix and corresponding clusters are uniquely defined. We apply LOVE to biological datasets (gene expression, serological responses measured from HIV controllers and chronic progressors, vaccine-induced humoral immune responses) resulting in meaningful biological output. For all three datasets, the clusters generated by LOVE remain stable across tuning parameters. Finally, we compared LOVE's performance to that of 13 state-of-the-art methods using previously established benchmarks and found that LOVE outperformed these methods across datasets. Our results demonstrate that LOVE can be broadly used across large-scale biological datasets to generate accurate and meaningful overlapping and non-overlapping clusters.  more » « less
Award ID(s):
1712709
PAR ID:
10097554
Author(s) / Creator(s):
; ; ;
Date Published:
Journal Name:
iScience
Volume:
14
ISSN:
2589-0042
Page Range / eLocation ID:
125-135
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; (, Biostatistics)
    null (Ed.)
    Summary High-throughput sequencing technology provides unprecedented opportunities to quantitatively explore human gut microbiome and its relation to diseases. Microbiome data are compositional, sparse, noisy, and heterogeneous, which pose serious challenges for statistical modeling. We propose an identifiable Bayesian multinomial matrix factorization model to infer overlapping clusters on both microbes and hosts. The proposed method represents the observed over-dispersed zero-inflated count matrix as Dirichlet-multinomial mixtures on which latent cluster structures are built hierarchically. Under the Bayesian framework, the number of clusters is automatically determined and available information from a taxonomic rank tree of microbes is naturally incorporated, which greatly improves the interpretability of our findings. We demonstrate the utility of the proposed approach by comparing to alternative methods in simulations. An application to a human gut microbiome data set involving patients with inflammatory bowel disease reveals interesting clusters, which contain bacteria families Bacteroidaceae, Bifidobacteriaceae, Enterobacteriaceae, Fusobacteriaceae, Lachnospiraceae, Ruminococcaceae, Pasteurellaceae, and Porphyromonadaceae that are known to be related to the inflammatory bowel disease and its subtypes according to biological literature. Our findings can help generate potential hypotheses for future investigation of the heterogeneity of the human gut microbiome. 
    more » « less
  2. (, Proceedings ICDM workshops)
    We have developed a Temporal Clustering with External Memory Network (TC-EMNet) for disease progression modeling that groups patients with similar trajectories to form disease clusters/stages. Experiments on two real-world datasets show that our model demonstrates competitive clustering performance against state-of-the-art methods and is able to identify clinically meaningful clusters. The visualization of the extracted patient states shows that the proposed model can generate better patient states than the baselines. 
    more » « less
  3. ; ; ; ; ; ; (, Bioinformatics)
    Abstract MotivationPredictive biological signatures provide utility as biomarkers for disease diagnosis and prognosis, as well as prediction of responses to vaccination or therapy. These signatures are identified from high-throughput profiling assays through a combination of dimensionality reduction and machine learning techniques. The genes, proteins, metabolites, and other biological analytes that compose signatures also generate hypotheses on the underlying mechanisms driving biological responses, thus improving biological understanding. Dimensionality reduction is a critical step in signature discovery to address the large number of analytes in omics datasets, especially for multi-omics profiling studies with tens of thousands of measurements. Latent factor models, which can account for the structural heterogeneity across diverse assays, effectively integrate multi-omics data and reduce dimensionality to a small number of factors that capture correlations and associations among measurements. These factors provide biologically interpretable features for predictive modeling. However, multi-omics integration and predictive modeling are generally performed independently in sequential steps, leading to suboptimal factor construction. Combining these steps can yield better multi-omics signatures that are more predictive while still being biologically meaningful. ResultsWe developed a supervised variational Bayesian factor model that extracts multi-omics signatures from high-throughput profiling datasets that can span multiple data types. Signature-based multiPle-omics intEgration via lAtent factoRs (SPEAR) adaptively determines factor rank, emphasis on factor structure, data relevance and feature sparsity. The method improves the reconstruction of underlying factors in synthetic examples and prediction accuracy of coronavirus disease 2019 severity and breast cancer tumor subtypes. Availability and implementationSPEAR is a publicly available R-package hosted at https://bitbucket.org/kleinstein/SPEAR. 
    more » « less
  4. ; ; ; (, Systematic Biology)
    Abstract Methods for rapidly inferring the evolutionary history of species or populations with genome-wide data are progressing, but computational constraints still limit our abilities in this area. We developed an alignment-free method to infer genome-wide phylogenies and implemented it in the Python package TopicContml. The method uses probabilistic topic modeling (specifically, Latent Dirichlet Allocation) to extract topic frequencies from k-mers, which are derived from multilocus DNA sequences. These extracted frequencies then serve as an input for the program Contml in the PHYLIP package, which is used to generate a species tree. We evaluated the performance of TopicContml on simulated datasets with gaps and three biological datasets: 1) 14 DNA sequence loci from two Australian bird species distributed across nine populations, 2) 5162 loci from 80 mammal species, and 3) raw, unaligned, nonorthologous PacBio sequences from 12 bird species. We also assessed the uncertainty of the estimated relationships among clades using a bootstrap procedure. Our empirical results and simulated data suggest that our method is efficient and statistically robust. 
    more » « less
  5. ; (, Bioinformatics)
    Mathelier, Anthony (Ed.)
    Abstract Motivation Methods to model dynamic changes in gene expression at a genome-wide level are not currently sufficient for large (temporally rich or single-cell) datasets. Variational autoencoders offer means to characterize large datasets and have been used effectively to characterize features of single-cell datasets. Here, we extend these methods for use with gene expression time series data. Results We present RVAgene: a recurrent variational autoencoder to model gene expression dynamics. RVAgene learns to accurately and efficiently reconstruct temporal gene profiles. It also learns a low dimensional representation of the data via a recurrent encoder network that can be used for biological feature discovery, and from which we can generate new gene expression data by sampling the latent space. We test RVAgene on simulated and real biological datasets, including embryonic stem cell differentiation and kidney injury response dynamics. In all cases, RVAgene accurately reconstructed complex gene expression temporal profiles. Via cross validation, we show that a low-error latent space representation can be learnt using only a fraction of the data. Through clustering and gene ontology term enrichment analysis on the latent space, we demonstrate the potential of RVAgene for unsupervised discovery. In particular, RVAgene identifies new programs of shared gene regulation of Lox family genes in response to kidney injury. Availability and implementation All datasets analyzed in this manuscript are publicly available and have been published previously. RVAgene is available in Python, at GitHub: https://github.com/maclean-lab/RVAgene; Zenodo archive: http://doi.org/10.5281/zenodo.4271097. Supplementary information Supplementary data are available at Bioinformatics online. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email