deconvolution

MethylBERT: A Transformer-based model for read-level DNA methylation pattern identification and tumour deconvolution

MethylBERT is a Transformer-based model that identifies tumour-derived reads and estimates tumour purity from bulk samples, outperforming existing methods in accuracy.

MODE: high-resolution digital dissociation with deep multimodal autoencoder

MODE improves deconvolution by integrating multi-omic data through a multimodal autoencoder, outperforming existing methods in simulation and real data.

Differentiating BOLD and non-BOLD signals in fMRI time series using cross-cortical depth delay patterns

This paper introduces a method to distinguish BOLD and non-BOLD signals in fMRI by analyzing cross-cortical depth delay patterns, demonstrating its effectiveness with visual task data.

Ultra-deep long-read metagenomics captures diverse taxonomic and biosynthetic potential of soil microbes

This study uses ultra-deep metagenomic sequencing to analyze soil microbes, revealing extensive microbial diversity and biosynthetic potential previously unexplored.

A Generalist Intracortical Motor Decoder

The paper trains an autoregressive Transformer on neural spiking data to decode motor behavior, showing broad utility across tasks but highlighting limitations in scaling.

cfDecon: Accurate and Interpretable methylation-based cell type deconvolution for cell-free DNA

cfDecon improves cfDNA deconvolution accuracy and interpretability using deep learning, outperforming existing methods in simulation and clinical datasets.

Deep Learning based deconvolution methods: a systematic review

This paper reviews deep learning methods for cellular deconvolution, emphasizing the need for high-quality reference profiles and standardized approaches.

Nanopore sequencing from protozoa to phages: decoding biological information on a string of biochemical molecules into human-readable signals

This study evaluates nanopore sequencing's ability to classify and assemble hypervariable genomes from metagenomic samples, showing high accuracy in identifying Trypanosoma brucei in infected mouse blood.

devider: long-read reconstruction of many diverse haplotypes

divider reconstructs haplotypes from long-read sequencing data, achieving higher accuracy than existing methods on synthetic and real datasets.

Estimating the effect of tissue- and blood-derived cell reference matrices on deconvolving bulk transcriptomic datasets

This study evaluates the impact of tissue- and blood-derived reference matrices on deconvolving bulk transcriptomic datasets, finding tissue-derived matrices perform better for tissue samples but not for blood.

Delineating the Spatial Patterns of Cell Type-Specific Stemness Using a Statistical Deconvolution Model

The paper develops a statistical deconvolution model to estimate cell type-specific stemness in spatial transcriptomics data and applies it to breast cancer data.

Unifying DNA methylation-based in silico cell-type deconvolution with deconvMe

The paper introduces deconvMe, an R package for DNA methylation-based cell-type deconvolution, and compares its results to gene expression and experimental data.

Unveiling the Mystery of the Petroleum Genome Archive: Paleofauna Activities, Geological History, and aDNA Identification Methods in Nanyang Oilfield

This study extracts and classifies DNA fragments from petroleum samples, revealing genetic insights into ancient environments and human history, while challenging traditional aDNA methods.

Detecting genetically variant peptides in non-human samples

The paper detects human genetically variant peptides in non-human samples using proteomics data reanalysis.

NLSDeconv: an efficient cell-type deconvolution method for spatial transcriptomics data

NLSDeconv is a cell-type deconvolution method for spatial transcriptomics data that offers competitive performance and superior computational efficiency compared to existing methods.

Bridging species boundaries: eDNA and genetic analysis reveal hybrid settlement at the very end of American eel distribution

The paper uses eDNA and genetic analysis to detect American eel and hybrid eels in Greenland, extending their known range and identifying hybrids.

Paternal obesity alters the sperm epigenome and is associated with changes in the placental transcriptome and cellular composition

Paternal obesity alters sperm epigenetic marks at placenta-related genes and is linked to changes in placental gene expression and trophoblast cell composition.

Single cell data enables dissecting cell types present in bulk transcriptome data

The paper presents a pipeline using scRNA-seq data to deconvolute and estimate cell type fractions in bulk transcriptome data from organoids.

Deconvolution of omics data in Python with Deconomix -- cellular compositions, cell-type specific gene regulation, and background contributions

Deconomix deconvolves bulk transcriptomics data to estimate cell compositions, background contributions, and cell-type-specific gene regulation.

scMoE: single-cell mixture of experts for learning hierarchical, cell-type-specific, and interpretable representations from heterogeneous scRNA-seq data

scMoE learns interpretable cell-type-specific representations from heterogeneous scRNA-seq data, improving clustering and spatial deconvolution across multiple datasets.

Assessing the performance of current strain resolution tools on long-read metagenomes

This study evaluates three strain resolution tools on long-read metagenomes, finding TRACS excels in haplotype completeness and efficiency, while Strainy shows high haplotype accuracy.

A deep neural network to de-noise single-cell RNAsequencing data

ZiPo is a deep neural network that de-noises scRNA-seq data by modeling dropouts and improving performance through a scale-invariant loss term.

Cell type-specific weighting-factors to solve solid organs-specific limitations of single cell RNA-sequencing

The paper introduces cell type-specific weighting-factors to address limitations in scRNA-seq analysis of solid tissues, improving bulk RNA-seq reconstitution and deconvolution accuracy.

Enhancing information extraction from field potentials in electrophysiology studies

This study introduces a marked point process representation of oscillatory bursts to reduce volume conduction noise in multi-channel neural recordings, preserving behavioral and causal information.

Investigating the Role of Management Decisions in Subspecies Hybridization Across the Wild Turkey's Range

This study examines how management decisions affect subspecies hybridization in wild turkeys, finding that introducing more subspecies and unique source states increases admixture, but not the number of relocated counties.

Characterization and Mitigation of a Simultaneous Multi-Slice fMRI Artifact: Multiband Artifact Regression in Simultaneous Slices

This paper identifies and corrects a novel artifact in multiband fMRI data using MARSS, improving signal-to-noise ratio and reducing systematic spatial patterns.

A brain-to-text framework of decoding natural tonal sentences

This paper develops a brain-to-text framework for decoding natural Mandarin tonal sentences from neural recordings, achieving higher accuracy than previous methods.

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.