Rafael A. Irizarry

Pan-cancer estimation of prognostic immune biomarkers from histopathology using artificial intelligence

This paper develops an AI model to estimate immune biomarkers from H&E images, showing it can predict prognostic outcomes as effectively as costly mIF assays.

Unique molecular identifiers don't need to be unique: a collision-aware estimator for RNA-seq quantification

The paper introduces a collision-aware estimator for RNA-seq quantification that allows shorter UMIs without sacrificing accuracy.

Spurious correlation inflates performance in single-cell perturbation prediction

The paper shows that performance metrics in single-cell perturbation prediction are inflated by reusing control cells, leading to overestimation of method effectiveness.

An image-based transcriptomics atlas reveals the regional and microbiota-dependent molecular, cellular, and spatial structure of the murine gut

This paper creates a gut transcriptomics atlas using MERFISH to show how microbiota influences cell types, spatial organization, and molecular responses in the murine gut.

Identifying spatially variable genes by projecting to morphologically relevant curves

The paper introduces a method to identify spatially variable genes by projecting spatial transcriptomics data onto morphologically relevant curves, improving detection accuracy and biological interpretation.

Robust identification of perturbed cell types in single-cell RNA-seq data

The paper introduces scDist, a tool for detecting perturbed cell types in single-cell RNA-seq data, showing it outperforms existing methods in identifying transcriptomic changes in disease contexts.

DNA binding analysis of rare variants in homeodomains reveals novel homeodomain specificity-determining residues

The paper analyzes DNA binding effects of 92 human homeodomain mutants, identifying 14 novel specificity-determining residues and highlighting limitations of variant effect prediction tools.

Significance Analysis for Clustering with Single-Cell RNA-Sequencing Data

The paper introduces a statistical method to assess the significance of clusters in single-cell RNA-seq data, improving accuracy by addressing variability and reducing false discoveries.

Cell type-specific inference of differential expression in spatial transcriptomics

The paper introduces a statistical method to detect cell type-specific differential gene expression in spatial transcriptomics data and validates it using simulations and real datasets.

Differential richness inference for 16S rRNA marker gene surveys

The paper introduces a method to infer differential richness in 16S rRNA surveys by accounting for spurious taxa generated during sequencing, using an R package called Prokounter.

A probabilistic gene expression barcode for annotation of cell-types from single cell RNA-seq data

This paper introduces a probabilistic method for cell-type annotation in scRNA-seq data that outperforms existing approaches by combining gene expression across thousands of genes and accounting for batch effects.

Detection of allele-specific expression in spatial transcriptomics with spASE

The paper introduces spASE, a method to detect spatial patterns in allele-specific expression from spatial transcriptomics data, showing it can identify cell type-specific effects in mouse hippocampus.

Vaginal Microbiome Topic Modelling of Laboring Ugandan Women With and Without Fever

This study uses microbial topic modeling to identify associations between vaginal microbiome composition and intrapartum fever in Ugandan laboring women, improving prediction beyond clinical variables.

Robust decomposition of cell type mixtures in spatial transcriptomics

This paper introduces RCTD, a method to decompose cell type mixtures in spatial transcriptomics, improving cell type assignment and enabling discovery of spatially dependent gene expression.

Quantile normalization of single-cell RNA-seq read counts without unique molecular identifiers

This paper introduces quasi-UMIs for normalizing scRNA-seq data without UMIs, improving accuracy over existing methods.

Feature Selection and Dimension Reduction for Single Cell RNA-Seq based on a Multinomial Model

The paper introduces multinomial-based methods for feature selection and dimension reduction in scRNA-Seq, improving clustering performance over existing approaches.

A gene expression atlas of embryonic neurogenesis in Drosophila reveals complex spatiotemporal regulation of lncRNAs.

The paper creates a spatiotemporal gene expression atlas of Drosophila embryonic neurogenesis, revealing dynamic lncRNA expression patterns and cell type-specific gene regulation.

Genome-wide repressive capacity of promoter DNA methylation is revealed through epigenomic manipulation

The paper reanalyzes data showing that forced DNA methylation of promoters strongly represses gene expression and reduces active transcription marks.

Selection Corrected Statistical Inference for Region Detection with High-throughput Assays

The paper presents a statistical method to correct selection bias in region detection from high-throughput assays, improving population inference and comparison of regions with different properties.

Detection and accurate False Discovery Rate control of differentially methylated regions from Whole Genome Bisulfite Sequencing

The paper introduces a method to detect differentially methylated regions from whole genome bisulfite sequencing data while accurately controlling the False Discovery Rate.

High-throughput identification of RNA nuclear enrichment sequences

The paper identifies RNA sequences responsible for nuclear enrichment in lncRNAs using a modified MPRA approach and validates findings with smRNA-FISH.

Technology-independent estimation of cell type composition using differentially methylated regions

The paper introduces a method to estimate cell type composition from DNA methylation data without relying on platform-specific reference profiles.

Accounting for GC-content bias reduces systematic errors and batch effects in ChIP-Seq data

The paper introduces a statistical method to account for GC-content bias in ChIP-Seq data, reducing false positives and improving cross-laboratory consistency.

Missing Data and Technical Variability in Single-Cell RNA- Sequencing Experiments

The paper examines technical variability and missing data in scRNA-seq, showing that technical zeros and cell-to-cell variation can confound biological findings.

Varying-Censoring Aware Matrix Factorization for Single Cell RNA-Sequencing

This paper introduces a matrix factorization method that accounts for censored data in scRNA-Seq to improve clustering accuracy.

Smooth Quantile Normalization

This paper introduces a normalization method for genomic data that accounts for biological differences in distributions while reducing technical variation.

Challenges And Emerging Directions In Single-Cell Analysis

This paper discusses challenges and future directions in single-cell analysis, highlighting advances in technologies and computational methods for studying cellular heterogeneity.

Correcting For Cell-Type Heterogeneity In Epigenome-Wide Association Studies: Premature Analyses And Conclusions

This paper critiques a study claiming ReFACTor improves EWAS by correcting cell-type heterogeneity, finding its claims lack rigorous statistical support.

Salmon provides accurate, fast, and bias-aware transcript expression estimates using dual-phase inference

Salmon provides accurate and fast transcript expression estimates by modeling and correcting GC content bias using a dual-phase inference algorithm.

When to use Quantile Normalization?

The paper introduces a data-driven method to assess the appropriateness of global normalization techniques like quantile normalization in gene expression and DNA methylation data.