deconvolution

Systematic errors in enzymatic conversion limit cell-free DNA methylation specificity

The paper identifies a systematic over-conversion error in EM-seq that introduces false-positive methylation signals in cfDNA analysis.

HEDeST: An Integrative Approach to Enhance Spatial Transcriptomic Deconvolution with Histology

HEDeST improves spatial transcriptomic deconvolution by integrating histology features, achieving single-cell resolution and outperforming existing methods in simulated and real cancer datasets.

S-MiXcan: Inferring Cell-Type-Level Transcriptome-Wide Associations from Bulk Transcriptomics Using GWAS Summary Statistics

Cell-type-aware TWAS from bulk-trained expression models and GWAS summary statistics reproduces individual-level MiXcan results (r > 0.99) and finds 32 genome-wide significant breast cancer genes.

A Zero-Inflated Hierarchical Generalized Transformation Model to Address Non-Normality in Spatially-Informed Cell-Type Deconvolution

This paper develops a zero-inflated hierarchical generalized transformation model to improve cell-type deconvolution in spatial transcriptomics data from oral squamous cell carcinomas.

SNMF: Ultrafast, Spatially-Aware Deconvolution for Spatial Transcriptomics

SNMF is a reference-free deconvolution method for spatial transcriptomics that improves accuracy and speed by incorporating spatial correlations and GPU acceleration.

Decoding real world visual scenes from the human gamma band with flicker evoked oscillations.

This paper decodes real-world visual scenes from human gamma band SSVEPs using flicker evoked oscillations, showing high accuracy with short data and highlighting the importance of 40 Hz.

Benchmarking tissue- and cell type-of-origin deconvolution in cell-free transcriptomics

This study benchmarks deconvolution methods for inferring tissue and cell type-of-origin from plasma cfRNA, showing method and reference variability impact performance.

Fixative eXchange (FX)-seq: Scalable Single-nucleus RNA Sequencing Analysis of PFA-fixed or FFPE Tissue

FX-seq improves snRNA-seq from FFPE and PFA-fixed tissues by removing crosslinks and preventing RNA leakage, enabling analysis of 321,710 nuclei from mouse and human samples.

Gradient boosting regression and convolution improve deconvolution of bulk transcriptomes

GrooD improves bulk transcriptome deconvolution using gradient boosting and convolution, showing better performance and interpretability on human blood samples from sepsis patients.

BEEP Learning: Multi-View Image Decomposition for Massively Multiplexed Biological Fluorescence Microscopy

This paper introduces BEEP learning, a machine learning framework for multi-view fluorescence unmixing that improves discrimination of fluorophores in multiplexed imaging.

In silico degradomics reveals disease- and endotype-specific alterations in the joint tissue landscape

The paper develops an in-silico pipeline to analyze degradomes from proteomic datasets, revealing disease- and endotype-specific tissue remodeling patterns in arthritis.

Spatiotemporal cell type deconvolution leveraging tissue structure

SpaDecoder improves cell type deconvolution in spatial transcriptomics by leveraging 3D tissue structure and single-cell reference profiles.

Resolving and Quantifying Viral-Like Particles via Blind Deconvolution

This paper introduces EpiVirQuant, a computer vision framework for automated enumeration and sizing of viral-like particles, showing 18% higher detection than manual counting.

Revisiting the cancer microbiome using PRISM

PRISM improves microbial analysis in low-biomass sequencing data, identifies tumor-associated microbes in cancer datasets, and reveals potential clinical significance.

Prion Seeding Activity in DNA Extractions: Implications for Laboratory Biosafety

The paper investigates whether DNA extraction methods co-purify prions, finding that commercial kits do not eliminate prion seeding activity, posing biosafety risks.

NanoSimFormer: An end-to-end Transformer-based simulator for nanopore sequencing signal data

NanoSimFormer generates high-fidelity nanopore sequencing signals optimized for accurate basecalling, outperforming existing simulators across multiple datasets and tasks.

High-fidelity bioimage restoration via adversarial learning

This paper introduces NAFNet GAN, a bioimage restoration framework that achieves high fidelity and low latency, suitable for real-time microscopy applications.

Mitigation and detection of putative microbial contaminant reads from long-read metagenomic datasets

This paper introduces a decontamination method for long-read metagenomic datasets, identifying and mitigating microbial contaminants in neonatal samples.

Integration of Imaging-based and Sequencing-based Spatial Omics Mapping on the Same Tissue Section via DBiTplus

Thermostable RNaseH release of barcoded cDNA keeps sections intact for CODEX imaging, yielding same-section whole-transcriptome and protein maps plus CODEX-guided spot deconvolution in FFPE tissues.

Non-negative matrix factorization and deconvolution as dual simplex problem

This paper reformulates non-negative matrix factorization as a dual simplex problem, improving decomposition accuracy in both image and biological data.

Benchmarking cell type deconvolution in spatial transcriptomics and application to cancer immunotherapy

This paper benchmarks cell type deconvolution methods in spatial transcriptomics, finding that a simple marker gene scoring approach performs well, especially for rare cell types, and applies it to study anti-PD1 immunotherapy in a mouse model.

Real-paired single-cell/bulk RNA-seq benchmark and a practical protocol for accurate cell-type deconvolution in human BAL samples

Systematic benchmarking reveals pseudo-bulk biases and provides practical fixes for accurate RNA deconvolution.

Denoising, Deblurring, and optical Deconvolution for cryo-ET and light microscopy with a physics-informed deep neural network DeBCR

DeBCR improves microscopy image quality through physics-informed deep learning, outperforming existing models in denoising, deblurring, and optical deconvolution across light microscopy and cryo-ET.

subsppLabelR: a wrapper package in R to automatically label and filter subspecies occurrence data

subsppLabelR automatically labels and filters subspecies occurrence data, performing well for well-sampled subspecies but struggling with rare or endemic ones.

HONeD-in on Brain Activity: Deconvolving Passive Diffusion on the Structural Network from Functional Brain Signals

The paper introduces HONeD-in, a method to deconvolve passive diffusion effects from functional MRI signals, revealing active brain activity patterns.

Immunoglobulin G binding as a quantitative marker of hepatocellular death across acute and chronic liver injury

Endogenous immunoglobulin G (IgG) immunostaining quantitatively labels apoptotic and necrotic hepatocytes across acute and chronic murine liver injury models, outperforming serum transaminases and histopathological Ishak scoring.

Modanovo: A Unified Model for Post-Translational Modification-Aware de Novo Sequencing Using Experimental Spectra from In Vivo and Synthetic Peptides

Modanovo is a de novo peptide sequencing model that handles post-translational modifications, outperforming existing methods and complementing database searches in phosphoproteomics.

'Guess Who?': Forensic genetics and archaeology converge to identify Cristopher Columbus descendants.

This study uses forensic genetic techniques to identify Columbus descendants from archaeological remains through familial relationship analysis.

Cellular deconvolution of the brain with topological magnetic resonance image analysis

The paper introduces STRATA, a machine learning framework that estimates cell type proportions from MRI data, validated across disease models and species.

Joint imputation and deconvolution of gene expression across spatial transcriptomics platforms

This paper introduces SIID, an algorithm that integrates spatial transcriptomics data from different platforms to impute missing gene expression and deconvolve mixed cell-type signals.