variational autoencoder

Building a learnable universal coordinate system for single-cell atlas with a joint-VAE model

The paper introduces UniCoord, a joint-VAE model that learns a universal coordinate system for single-cell transcriptomic data to capture heterogeneity and enable data generation and analysis.

Adversarial learning enables unbiased organism-wide cross-species alignment of single-cell RNA data at scale

This paper introduces a method to align single-cell RNA data across species and tissues using adversarial learning, revealing common cell types like macrophages and fibroblasts.

Disentangling Cobionts and Contamination in Long-Read Genomic Data using Sequence Composition

The paper develops a method to identify cobionts, parasites, and contaminants in long-read genomic data using sequence composition and a Variational Autoencoder, with minimal reliance on reference data.

A deep generative model for capturing cell to phenotype relationships

The paper introduces milVI, a deep generative model that infers missing phenotype labels and connects cell-level data to donor-level phenotypes in single-cell studies.

scCross: A Deep Generative Model for Unifying Single-cell Multi-omics with Seamless Integration, Cross-modal Generation, and In-silico Exploration

scCross integrates and generates single-cell multi-omics data using deep generative models, enabling seamless cross-modal analysis and in-silico experiments.

Single-cell multi-omics and spatial multi-omics data integration via dual-path graph attention auto-encoder

This paper introduces SSGATE, a dual-path graph attention auto-encoder for integrating single-cell and spatial multi-omics data, showing improved performance and robustness over existing methods.

A Denoising Convolutional Autoencoder for SNR Enhancement in Chemical Exchange Saturation Transfer imaging: (DCAE-CEST)

This paper develops a denoising convolutional autoencoder for CEST imaging to enhance SNR and compares it to existing methods, showing improved performance in both simulated and in vivo data.

Deep Normative Tractometry for Identifying Joint White Matter Macro- and Micro-structural Abnormalities in Alzheimer's Disease

This paper introduces DNT, a variational autoencoder framework for detecting white matter abnormalities in Alzheimer's disease using macro- and microstructural data.

Jointly efficient encoding and decoding in neural populations

This paper introduces a framework combining encoding and decoding in neural populations using variational autoencoders, showing that models with broad tuning curves achieve accurate stimulus reconstruction and generalization.

Unsupervised representation learning improves genomic discovery and risk prediction for respiratory and circulatory functions and diseases

This paper introduces a deep learning framework, REGLE, that improves genomic discovery and risk prediction for respiratory and circulatory diseases by learning low-dimensional embeddings from high-dimensional clinical data.

Interpretable variational encoding of genotypes identifiescomprehensive clonality and lineages in single cells geometrically

SNPmanifold uses a variational autoencoder to identify clonality and lineages in single cells, outperforming existing methods in complex datasets.

A variational autoencoder trained with priors from canonical pathways increases the interpretability of transcriptome data

This study introduces a prior-informed VAE for transcriptome data that improves interpretability by learning latent features aligned with biological pathways.

Imputation of label-free quantitative mass spectrometry-based proteomics data using self-supervised deep learning

This paper introduces a deep learning method for imputing missing values in label-free quantitative proteomics data and shows its effectiveness in identifying differentially abundant proteins in alcohol-related liver disease.

In silico generation and augmentation of regulatory variants from massively parallel reporter assay using conditional variational autoencoder

The paper introduces MpraVAE, a deep generative model that generates and augments regulatory variant data from MPRA, improving prediction performance for regulatory variants in autoimmune diseases.

Predictive Sequence Learning in the Hippocampal Formation

The paper develops a predictive autoencoder model of the hippocampus to simulate sequence learning, showing CA3's predictive role and CA1's error signaling in mice.

AutoRNA: RNA tertiary structure prediction using variational autoencoder.

AutoRNA uses a variational autoencoder to predict RNA tertiary structures from sequences, achieving reasonable accuracy but limited global fold recovery.

scNODE: Generative Model for Temporal Single Cell Transcriptomic Data Prediction

scNODE predicts single-cell gene expression at unobserved timepoints using a deep learning model combining VAE and ODEs, improving trajectory inference and perturbation analysis.

Low-dimensional representations of genome-scale metabolism

This paper uses self-supervised learning to create low-dimensional representations of genome-scale metabolic flux data, improving cell type classification and comparison across organisms.

Improving allele-specific epigenomic signal coverage by 10-fold using Hidden Markov Modeling and Machine Learning

This paper improves allele-specific epigenomic signal coverage by 10-fold using variational hidden Markov modeling and machine learning, demonstrating it on DNA methylation data.

Designing meaningful continuous representations of T cell receptor sequences with deep generative models

This paper introduces TCR-VALID, a variational autoencoder for generating and analyzing TCR sequences, showing improved clustering and representation quality.

Explaining Deep Learning-Based Representations of Resting State Functional Connectivity Data: Focusing on Interpreting Nonlinear Patterns in Autism Spectrum Disorder

This paper introduces explainable variational autoencoders to interpret nonlinear patterns in resting state fMRI data from autism spectrum disorder and control groups.

Generating Synthetic Single Cell Data from Bulk RNA-seq Using a Pretrained Variational Autoencoder

This study develops a variational autoencoder to generate synthetic scRNA-seq data from bulk RNA-seq, enabling single-cell insights from large-scale bulk datasets.

Computing Linkage Disequilibrium Aware Genome Embeddings using Autoencoders

The paper introduces a method to compress genetic data using autoencoders that leverage linkage disequilibrium structure, achieving high reconstruction accuracy while reducing data size.

A Demographic-Conditioned Variational Autoencoder for fMRI Distribution Sampling and Removal of Confounds

This paper introduces DemoVAE, a variational autoencoder that generates synthetic fMRI data conditioned on demographics and removes demographic confounds from fMRI features.

Identifying mechanisms and therapeutic targets in muscle using Bayesian parameter estimation with conditional variational autoencoders

This paper uses a Conditional Variational Autoencoder to predict muscle twitch parameters from simulated and experimental data, identifying therapeutic targets for a cardiac mutation.

Scalable Integration of Multiomic Single Cell Data Using Generative Adversarial Networks

This paper introduces a framework for integrating multiomic single-cell data using Wasserstein GANs, enabling analysis of more than two modalities with reduced computational burden.

Variational Bayesian Phylogenies through Matrix Representation of Tree Space

The paper introduces a matrix representation of tree space for Bayesian phylogenetic inference, demonstrating it can recover tree parameters and produce better summary trees than traditional methods.

ISMI-VAE: A Deep Learning Model for Classifying Disease Cells Using Gene Expression and SNV Data

This paper introduces ISMI-VAE, a deep learning model that integrates SNV and gene expression data to classify disease cells, showing improved effectiveness and interpretability.

STEP: Spatial Transcriptomics Embedding Procedure for Multi-scale Biological Heterogeneities Revelation in Multiple Samples

STEP is a novel AI architecture for spatial transcriptomics data that improves the identification of cell types and spatial domains, with enhanced accuracy in liver zone demarcation.

Latent-based Directed Evolution accelerated by Gradient Ascent for Protein Sequence Design

This paper introduces LDE, a variational autoencoder-based method for protein sequence design that improves efficiency by prioritizing high-fitness mutants in latent space, outperforming baselines on eight tasks.