Yoseph Barash

This paper introduces TrASPr+BOS, a generative AI model for predicting and designing RNA to achieve tissue-specific splicing outcomes, showing improved performance over existing methods.

The paper identifies sequence features associated with recursive splicing and develops a classifier to detect these features.

The paper investigates how RNA G-quadruplexes enable cooperative binding of HNRNPH and regulate splicing, showing clinical relevance in breast cancer.

G4mer identifies G-quadruplexes and disease variants from population-scale genetic data, improving accuracy and highlighting sequence features affecting rG4 formation.

The paper improves sQTL modeling by incorporating alternative splicing representation and introduces MAJIQTL, finding new sQTLs linked to Alzheimer's disease.

The paper identifies DDX55 as a novel regulator of 3'UTR isoform diversity through integrative analysis of RNA binding protein depletion experiments.

The paper identifies a NRCAM proteoform with skipped microexons as a targetable surface antigen in pHGG and demonstrates its utility in immunotherapy.

MAJIQ-CLIN improves the detection of Mendelian disease-causing variants from RNA-Seq by addressing splicing aberrations, showing better accuracy and efficiency than existing tools.

This paper introduces LSV-seq, a targeted RNA-seq method optimized with machine learning to detect alternative splicing with high sensitivity and low sequencing depth.

The paper compares short and long read RNA sequencing technologies, finds they detect different transcriptome features, and introduces software to combine their results for improved analysis.

This study shows RBMX and RBMXL2 proteins regulate splicing of ultra-long exons by repressing cryptic splice sites, with RBMXL2 rescuing splicing defects in somatic cells.

The paper benchmarks eight tools for identifying and quantifying alternative polyadenylation sites from RNA-seq data using real, synthetic, and 3'-end sequencing data.

The paper introduces MAJIQ v2, a tool for analyzing RNA splicing in large and heterogeneous RNA-seq datasets, showing improved performance over existing methods and applying it to brain subregion-specific splicing analysis.

The paper identifies mutations and RNA-binding proteins affecting CD19 splicing and CART-19 therapy resistance through high-throughput mutagenesis and mathematical modeling.

MOCCASIN corrects known and unknown confounders in RNA splicing analysis using synthetic and real data from TARGET and ENCODE.

The paper identifies ESRP1's genome-wide binding sites and shows its role in splicing regulation and post-transcriptional functions in epithelial cells.

The paper shows that alternative splicing contributes to gene disruption in AML, increasing the number of patients with functional gene loss beyond what is detected by mutations alone.

The paper identifies uORF variants that disrupt upstream translation and are under strong negative selection, linking them to gene-disease associations and loss-of-function in human genes.

The paper introduces fastGRO, a method for rapid and scalable profiling of nascent RNA, showing it is more comprehensive and scalable than existing techniques.

The paper shows RBMX prevents non-productive splicing of ultra-long exons, maintaining genome stability by blocking aberrant splice sites in genes like ETAA1.

The paper assesses how RNA-seq from clinically-accessible tissues represents splicing in non-accessible tissues and introduces a resource to explore these findings.

The paper identifies G-quadruplex sequences in UTRs as functionally constrained and enriched for cis-eQTLs and RNA-binding protein interactions, linking them to disease associations.

The paper extends the Integrated Gradients method for better interpretability of deep learning models in genomics by incorporating nonlinear paths and alternative baselines.

The paper critiques Li et al.'s comparative analysis of splicing detection tools, highlighting issues with the representation of MAJIQ and rMATS.

This study identifies A1CF as a regulator of plasma triglycerides through posttranscriptional control of VLDL secretion and intracellular stress pathways.

The paper develops a probability model to detect outlier RNA-Seq samples for improved differential splicing analysis, showing it outperforms existing tools in identifying inconsistent replicates.

MAJIQ-SPEL is a web-tool that visualizes and quantifies gene isoforms from RNA-Seq data, handling both classical and complex splicing variations and suggesting primers for experimental validation.

The paper identifies antagonistic splicing regulation between CELF2 and RBFOX family proteins in human T cells and shows evolutionary conservation across species.

This paper improves AS prediction models by integrating diverse data sources and introduces a new target function that enhances model accuracy.