deconvolution

Self-inspired learning to denoise for live-cell super-resolution microscopy

This paper introduces SN2N, a self-supervised denoising method for live-cell super-resolution microscopy that achieves competitive performance without requiring large clean datasets.

GLDADec: marker-gene guided LDA modelling for bulk gene expression deconvolution

GLDADec deconvolves bulk gene expression data using marker gene names to estimate cell type proportions, outperforming existing methods and enabling clinical applications.

Pangenome databases provide superior host removal and mycobacteria classification from clinical metagenomic data

The paper benchmarks host removal and mycobacteria classification methods in metagenomic data, finding that custom databases improve accuracy and efficiency.

FONDUE: Robust resolution-invariant denoising of MR Images using Nested UNets

This paper introduces FONDUE, a deep learning method for denoising MRI images that works across varying resolutions and performs comparably to state-of-the-art methods.

CellsFromSpace: A fast, accurate and reference-free tool to deconvolve and annotate spatially distributed Omics data

CellsFromSpace is a reference-free tool for deconvolving and annotating spatial transcriptomic data, outperforming existing methods in accuracy and speed across multiple technologies.

CATD: A reproducible pipeline for selecting cell-type deconvolution methods across tissues

This paper evaluates 31 cell-type deconvolution methods across human and mouse tissues, identifies regression-based methods as effective, and introduces the CATD pipeline for reproducible analysis.

Deconer: A comprehensive and systematic evaluation toolkit for reference-based cell type deconvolution algorithms using gene expression data

Deconer evaluates reference-based cell type deconvolution methods using simulated and real gene expression data, comparing 14 methods across multiple criteria and providing guidance for users.

Smoother: A Unified and Modular Framework for Incorporating Structural Dependency in Spatial Omics Data

Smoother integrates positional information into non-spatial models for spatial omics data, enabling accurate imputation, deconvolution, and dimensionality reduction in simulated and real datasets.

Cellos: High-throughput deconvolution of 3D organoid dynamics at cellular resolution for cancer pharmacology

Cellos is a high-throughput image analysis pipeline for 3D organoid segmentation and analysis, enabling accurate quantification of drug responses and cell interactions in cancer organoids.

STIE: Single-cell level deconvolution, convolution, and clustering in spatial transcriptomics by aligning spot level transcriptome to nuclear morphology

Highly accurate estimation of cell type abundance in bulk tissues based on single-cell reference and domain adaptive matching

SCROAM accurately estimates cell type abundance in bulk tissues using single-cell references and domain adaptation, outperforming existing methods in multiple datasets.

A Pose-Informed De-Noising Diffusion Model for Adult Naturalistic EEG Signals

This paper introduces a denoising diffusion model for EEG signals that uses pose coordinates to improve denoising performance and maintain spatial dependencies.

Adaptive Regularized Tri-Factor Non-Negative Matrix Factorization for Cell Type Deconvolution

ARTdeConv improves cell type deconvolution by addressing limitations in existing methods, showing better performance in simulations and real data compared to other approaches.

DeconV: Probabilistic Cell Type Deconvolution from Bulk RNA-sequencing Data

DeconV is a probabilistic cell-type deconvolution method for bulk RNA-Seq data using scRNA-Seq references, offering comparable accuracy to Scaden with added interpretability through confidence intervals.

Deconvolution of Nascent Sequencing Data Using Transcriptional Regulatory Elements

This paper develops a method for supervised deconvolution of nascent sequencing data and finds enhancer RNAs are informative, while undifferentiated cell types complicate deconvolution.

Hostile: accurate host decontamination of microbial sequences

Hostile is a tool for removing human host sequences from microbial sequencing data, achieving high accuracy in read removal while preserving target microbial sequences.

Locally low-rank denoising in transform domains

This paper extends locally low-rank denoising techniques to transform domains, improving MR image denoising with fewer images.

Proteoform-level deconvolution reveals a broader spectrum of ibrutinib off-targets

This study identifies additional proteoform targets of ibrutinib using thermal proteome profiling, revealing off-target interactions linked to immunomodulation and cellular processes.

Fast alignment of mass spectra in large proteomics datasets, capturing dissimilarities arising from multiple complex modifications of peptides

SpecGlobX improves peptide-spectrum match alignment by handling complex modifications, enhancing identification confidence in proteomics data.

Rat Deconvolution as Knowledge Miner for Immune Cell Trafficking from Toxicogenomics Databases

This paper develops a rat-specific deconvolution method to retrieve immune cell trafficking information from toxicogenomics databases, finding species-specific differences and identifying compound clusters based on immune cell profiles.

Deep Learning-based Optical Aberration Estimation Enables Offline Digital Adaptive Optics and Super-resolution Imaging

This paper introduces SFE-Net and SFT-DFCAN to estimate and correct optical aberrations in live-cell imaging without additional hardware, achieving fast and accurate super-resolution.

SecDATA: Secure Data Access and de novo TranscriptAssembly protocol - To meet the challenge of reliableNGS data analysis

SecDATA provides a secure data access protocol with a graph-based de novo transcript assembly method, outperforming existing tools in transcript reconstruction.

Characterization and decontamination of background noise in droplet-based single-cell protein expression data with DecontPro

The paper identifies and removes background noise in CITE-seq data using DecontPro, a Bayesian model that improves clustering specificity and reduces contamination from spongelets and ambient ADTs.

Rational probe design for efficient rRNA depletion and improved metatranscriptomic analysis of human microbiomes

The paper introduces a probe set for efficient rRNA depletion in human microbiome samples, improving metatranscriptomic analysis and reducing quantitative error.

Deep Domain Adversarial Neural Network for the Deconvolution of Cell Type Mixtures in Tissue Proteome Profiling

This paper introduces scpDeconv, a deep learning method for deconvolving cell type mixtures in proteomic data, showing improved performance across species and technologies.

Large-scale evaluation of outcomes following a genetic diagnosis in children with severe developmental disorders

The paper evaluates clinical outcomes after genetic diagnosis in children with severe developmental disorders, finding that most families received condition-specific information and support.

Optimising Spectronaut search parameters to improve data quality with minimal proteome coverage reductions in DIA analyses of heterogeneous samples

The paper optimizes Spectronaut search parameters to reduce false positives in DIA proteomics with minimal loss of true positives in heterogeneous samples.

DeMixSC: a deconvolution framework that uses single-cell sequencing plus a small benchmark dataset for improved analysis of cell-type ratios in complex tissue samples

DeMixSC improves deconvolution accuracy by using single-cell RNA-seq and a benchmark dataset to account for technological discrepancies in complex tissue samples.

HoCoRT: Host contamination removal tool

HoCoRT removes host contamination from shotgun metagenome sequencing data using multiple methods, finding optimal combinations for accuracy and speed.

Deepurify: a multi-modal deep language model to remove contamination from metagenome-assembled genomes

Deepurify uses a multi-modal deep language model with contrastive learning to improve MAG decontamination by leveraging genomic sequences and taxonomic lineages.