deconvolution

ChromaFold predicts the 3D contact map from single-cell chromatin accessibility

ChromaFold predicts 3D chromatin contact maps from scATAC-seq data, outperforming bulk-based methods and enabling cell-type-specific interaction analysis.

Clustering-independent estimation of cell abundances in bulk tissues using single-cell RNA-seq data

The paper introduces ConDecon, a clustering-independent method to estimate cell abundances in bulk tissues using single-cell RNA-seq data, and applies it to study neurodegenerative microglia pathways in pediatric ependymoma.

PENGUIN: A rapid and efficient image preprocessing tool for multiplexed spatial proteomics

PENGUIN is a preprocessing tool for multiplexed spatial proteomics that normalizes signal-to-noise ratios and reduces noise without manual annotation or machine learning training.

Diffusion-based artificial genomes and their usefulness for local ancestry inference

This study introduces a frugal diffusion model for generating artificial genomes and evaluates its utility for local ancestry inference using synthetic and real data.

Physics-Informed Ellipsoidal Coordinate Encoding Implicit Neural Representation for high-resolution volumetric wide-field microscopy

This paper introduces PIECE-INR, a physics-informed method for high-fidelity 3D fluorescence data reconstruction in wide-field microscopy, achieving state-of-the-art results without requiring ground truth data.

DeepEXPOKE: A Deep Learning Framework with Polygenic Risk Scores as Knockoffs for Deconvoluting Genetic and Non-Genetic Exposure Risks in Sepsis and Coronary Heart Disease

DeepEXPOKE uses deep learning with knockoff features to distinguish genetic and non-genetic exposure risks in sepsis and coronary heart disease, identifying metabolites and asthma as risk factors.

Accurate Cell Abundance Quantification using Multi-positive and Unlabeled Self-learning

CleverXMBD1 improves cell abundance quantification by using a confidence-based multi-positive and unlabeled loss function, outperforming existing methods across multiple datasets.

Robust estimation of cancer and immune cell-type proportions from bulk tumor ATAC-Seq data.

This paper develops EPIC-ATAC to estimate cancer and immune cell-type proportions from bulk tumor ATAC-Seq data, showing accurate prediction of immune contexture in breast cancer subtypes.

DigestR an open-source software tool for visualizing LC-MS proteomics data resulting from natural protein catabolism

DigestR is an open-source tool for visualizing LC-MS proteomics data from natural protein catabolism, demonstrating its ability to identify proteolytic cut sites in human and parasite systems.

Increasing Representativeness in the All of Us Cohort Using Inverse Probability Weighting

The paper develops inverse probability weights to reduce volunteer bias in the All of Us cohort, improving its representativeness and external validity.

Informing biologically relevant signal from spatial transcriptomic data

The paper benchmarks deconvolution tools for spatial transcriptomic data and introduces a novel pipeline to analyze tissue microenvironment dynamics.

Community assessment of methods to deconvolve cellular composition from bulk gene expression

The paper evaluates deconvolution methods for inferring immune infiltration from bulk gene expression using a DREAM Challenge, finding that some community-contributed methods, especially deep learning-based ones, perform well.

MetDecode: methylation-based deconvolution of cell-free DNA for non-invasive multi-cancer typing

MetDecode uses methylation patterns to estimate tumor proportion and identify tissue origin in cell-free DNA samples from cancer patients.

A single-molecule nanopore sequencing platform

This paper introduces an improved nanopore sequencing platform called CycloneSEQ, demonstrating long-duration sequencing and high yield with applications in genomics.

A Biologically Inspired Attention Model for Neural Signal Analysis

This paper introduces SPARKS, a biologically inspired model that achieves high decoding accuracy and interpretability in neural signal analysis across multiple datasets.

Decoding dynamic visual scenes across the brain hierarchy

This study uses deep learning to decode dynamic natural visual scenes from neural activity across brain regions, revealing distinct encoding patterns in visual cortex and subcortical nuclei compared to hippocampus.

BLEND: Probabilistic Cellular Deconvolution with Automated Reference Selection

BLEND improves cellular deconvolution by selecting optimal references and avoiding data preprocessing, showing better performance on real and simulated data.

Extending the capabilities of deconvolution to provide cell type specific pathway analysis of bulk RNA-seq data for idiopathic pulmonary fibrosis

This paper applies deconvolution to bulk RNA-seq data from idiopathic pulmonary fibrosis to identify cell-type specific pathway changes.

Sequence-to-sequence translation from mass spectra to peptides with a transformer model

Casanovo translates tandem mass spectra into peptide sequences using a transformer model, outperforming existing methods and improving analysis in immunopeptidomics and metaproteomics.

Unlocking Antarctic molecular time-capsules - recovering historical environmental DNA from museum-preserved sponges

The paper recovers historical environmental DNA (heDNA) from Antarctic sponge specimens to reconstruct past fish biodiversity and community composition.

InstaPrism: an R package for fast implementation of BayesPrism

InstaPrism re-implements BayesPrism with a fixed-point algorithm to improve computational efficiency while maintaining accuracy in cell-type deconvolution.

sNucConv: A bulk RNA-seq deconvolution method trained on single-nucleus RNA-seq data to estimate cell-type composition of human subcutaneous and visceral adipose tissues

QRATER: a collaborative and centralized imaging quality control web-based application.

QRATER is a web-based QC tool for neuroimaging that enables rating of QC images, showing varying time and agreement across tasks and rater experience.

An amplicon panel for high-throughput and low-cost genotyping of Pacific oyster

The paper develops an amplicon panel for genotyping Pacific oysters to support breeding programs and identifies null alleles affecting parentage analysis.

Decoding movie content from neuronal population activity in the human medial temporal lobe

The paper decodes movie features like characters, settings, and visual transitions from neuronal population activity in the human medial temporal lobe, showing region-specific information encoding.

Enhancing the biological insight from limited proteolysis-coupled mass spectrometry

The paper introduces a framework to deconvolve LiP-MS signals and infer protein structural changes using data from three species.

Multi-cell type deconvolution using a probabilistic model for single-molecule DNA methylation haplotypes

This paper develops a probabilistic model for cell type deconvolution using single-molecule DNA methylation haplotypes, showing improved accuracy in detecting rare cell types.

Fine-scale cellular deconvolution via generalized maximum entropy on canonical correlation features

This paper introduces a method for fine-scale cellular deconvolution using generalized maximum entropy on canonical correlation features, achieving higher resolution and computational efficiency than existing approaches.

Identification of 5mC within heterogenous tissue using de-novo somatic mutations

The paper uses de-novo somatic mutations to deconvolute 5mC methylation patterns in mixed tissues, showing improved accuracy in estimating methylation levels and detecting differentially-methylated sites.

A transformer model for de novo sequencing of data-independent acquisition mass spectrometry data

This paper introduces a transformer-based model called Cascadia for de novo sequencing of data-independent acquisition mass spectrometry data, achieving state-of-the-art performance.