deconvolution

ProteinSGM: Score-based generative modeling for de novo protein design

This paper introduces ProteinSGM, a score-based diffusion model for de novo protein design, showing it can generate native-like structures and perform conditional design.

Investigation of the usefulness of liver-specific deconvolution method toward legacy data utilization

The paper evaluates a liver-specific deconvolution method on legacy data, showing improved extrapolation when using optimized reference cell sets for liver transcriptome analysis.

Sparsity of higher-order landscape interactions enables learning and prediction for microbiomes

This paper introduces a mechanism-agnostic method to predict microbial community compositions and functions using sparse representations and compressive sensing, validated on both simulated and experimental data.

SpatialCTD: a large-scale TME spatial transcriptomic dataset to evaluate cell type deconvolution for immuno-oncology

The paper introduces a large-scale spatial transcriptomic dataset and a Graph Neural Network-based method for cell type deconvolution in human immuno-oncology.

SpaDecon: cell-type deconvolution in spatial transcriptomics with semi-supervised learning

SpaDecon improves cell-type deconvolution in spatial transcriptomics using semi-supervised learning, outperforming existing methods on real and pseudo datasets.

GEGVIC: A workflow to analyze Gene Expression, Genetic Variations and Immune cell Composition of tumor samples using Next Generation Sequencing data

GEGVIC is an R tool that simplifies cancer research analysis by integrating gene expression, genetic variation, and immune cell composition from sequencing data through a modular pipeline and GUI.

An emerging view of neural geometry in motor cortex supports high-performance decoding

This paper introduces MINT, a decoder for brain-computer interfaces that uses statistical constraints, outperforming interpretable methods and competing with machine learning approaches in multiple tasks.

Unmixing Fluorescence Image Data with Sparse and Low-Rank Poisson Regression

This paper introduces a sparse and low-rank Poisson unmixing method to improve fluorescence image unmixing accuracy in low signal-to-noise regimes with overlapping fluorophores.

DIP-MS: A novel ultra-deep interaction proteomics for the deconvolution of protein complexes

DIP-MS resolves protein complex isoforms using deep learning and mass spectrometry, revealing new subunits in human prefoldin complexes.

Spatial Transcriptomics Using Multiplexed Deterministic Barcoding in Tissue

This study introduces xDbit, a multiplexed spatial transcriptomics method, and demonstrates its application in characterizing murine kidney tissue.

ELITE: Expression deconvoLution using lInear optimizaTion in bulk transcriptomics mixturEs

ELITE deconvolves bulk transcriptomic data using linear programming to estimate cell type composition, outperforming existing methods in cases with similar expression profiles.

Noise-robust, physical microscopic deconvolution algorithm enabled by multi-resolution analysis regularization

This paper introduces noise-robust deconvolution algorithms for fluorescence microscopy using multi-resolution analysis regularization, improving resolution and SNR across various imaging techniques.

Spatially informed clustering, integration, and deconvolution of spatial transcriptomics with GraphST

GraphST improves spatial transcriptomics analysis by integrating clustering, integration, and deconvolution through graph self-supervised contrastive learning, achieving higher accuracy than existing methods.

Cell-type deconvolution of bulk RNA-Seq from kidney using opensource bioinformatic tools

This paper presents an open-source method to assess cell-type composition in bulk RNA-Seq data from heterogeneous tissues using scRNA-Seq references, showing no sex-based differences in kidney transcriptomes from baboons.

Spanve: A Statistical Method for Detecting Downstream-Friendly Spatially Variable Genes in Large-Scale Spatial Transcriptomic Data

Spanve is a nonparametric statistical method that identifies spatially variable genes in spatial transcriptomic data with fewer false positives, enabling better downstream analysis.

Deduplication Improves Cost-Efficiency and Yields of De novo Assembly and Binning of Shot-Gun Metagenomes in Microbiome Research

This study shows that removing duplicate reads improves metagenomic assembly and binning yields and reduces computational costs in complex environmental samples.

Multienzyme deep learning models improve peptide de novo sequencing by mass spectrometry proteomics

This study develops multienzyme deep learning models for de novo peptide sequencing from mass spectrometry data, showing improved generalizability and sequence coverage across diverse protease digests and species.

DESP: Demixing Cell State Profiles from Dynamic Bulk Molecular Measurements

DESP resolves cell state contributions to bulk molecular measurements using single-cell data, demonstrating accuracy in reconstructing proteome and transcriptome signatures during epithelial-to-mesenchymal transition.

Deep learning-driven fragment ion series classification enables highly precise and sensitive de novo peptide sequencing

This paper introduces Spectralis, a deep learning method for de novo peptide sequencing that improves sensitivity and precision in tandem mass spectrometry.

Spatial transcriptomics deconvolution at single-cell resolution by Redeconve

Redeconve deconvolutes spatial transcriptomics data at single-cell resolution, outperforming existing methods in accuracy and speed across multiple datasets.

Guided-topic modelling of single-cell transcriptomes enables sub-cell-type and disease-subtype deconvolution of bulk transcriptomes

This paper introduces GTM-decon, a method to deconvolve bulk transcriptomes into cell-type and disease-subtype compositions using single-cell RNA-seq data.

Accurate cell type deconvolution in spatial transcriptomics using a batch effect-free strategy

This paper introduces ReSort, a method to improve cell type deconvolution in spatial transcriptomics by reducing batch effects, and shows it enhances accuracy in mouse breast cancer data.

Visual QC Protocol for FreeSurfer Cortical Parcellations from Anatomical MRI

The paper introduces a visual quality control protocol for FreeSurfer cortical parcellations using VisualQC, showing it effectively detects errors and has reliable inter-rater agreement.

Cell composition inference and identification of layer-specific spatial transcriptional profiles with POLARIS

POLARIS infers cell composition from histology images and identifies layer-specific differentially expressed genes in spatial transcriptomics data.

wastewaterSPAdes: SARS-CoV-2 strain deconvolution using SPAdes toolkit

This paper introduces wastewaterSPAdes, an assembly-based method for SARS-CoV-2 strain deconvolution from wastewater samples without read alignment, showing accurate strain identification and abundance estimation.

Performance of computational algorithms to deconvolve heterogeneous bulk tumor tissue depends on experimental factors

This study evaluates how experimental factors affect the performance of tumor deconvolution algorithms using bulk RNA-seq data.

Computational deconvolution of fifteen leukocyte subtypes from DNA methylation microarrays trained on flow cytometry data in the Health and Retirement Study

The paper trains a computational model to deconvolve fifteen leukocyte subtypes from DNA methylation data using flow cytometry data, improving variance explanation and reducing false positives in age studies.

Deconvolution of plasma pharmacokinetics from dynamic heart imaging data obtained by SPECT/CT imaging

This paper develops a compartmental model to deconvolve plasma pharmacokinetics from dynamic heart imaging data, enabling accurate analysis of radiotracer kinetics without invasive plasma sampling.

The usage of transcriptomics datasets as sources of Real-World Data for clinical trialling

This paper constructs a patient repository by integrating transcriptomics datasets to generate real-world data for clinical trialling, improving downstream biological conclusions by 45%.

Deep autoencoder for interpretable tissue-adaptive deconvolution and cell-type-specific gene analysis

TAPE deconvolves bulk RNA-seq data into cell-type fractions and gene expression using a deep autoencoder, showing better performance and robustness compared to existing methods.