Páll Melsted

The paper uses population-scale Ig genetics to identify genetic associations and map their effects on the human B-cell system, revealing new genes and regulatory mechanisms in humoral immunity.

BloodVariome identifies 259 genetic associations across 127 immune cell populations in 11,983 individuals, revealing fine-grained genetic compartmentalization of the immune system.

This study compares Seurat and Scanpy, finding significant differences in their scRNA-seq analysis outputs, comparable to low sequencing depth or cell coverage.

The paper presents a GPU-accelerated version of kallisto for RNA-seq transcript quantification, achieving significant speedups over CPU implementations.

The paper develops lr-kallisto, a tool for accurate long-read transcriptome quantification using ONT data, showing improved results with exome capture.

The paper describes using kallisto, bustools, and kb-python to quantify RNA-seq data from bulk, single-cell, and single-nucleus samples.

This study evaluates batch correction methods for scRNA-seq and finds that most introduce artifacts, with Harmony being the only consistently reliable method.

Geospatial exploratory statistics, including local, bivariate and multivariate spatial methods, are packaged for spatial omics behind one interface, with matched R/Bioconductor and Python implementations.

The paper presents a high-quality chromosome-level genome assembly for the Rock Ptarmigan, providing a reference for studying its evolution and climate change impacts.

The paper introduces a compression algorithm for BUS files that achieves faster speeds and smaller file sizes than gzip.

The paper introduces a method to discover associations between novel gene isoforms and phenotypes without relying on existing annotations.

This paper introduces a method to improve RNA-seq quantification by expanding the transcriptome reference to include both nascent and mature mRNA and using distinguishing flanking k-mers to enhance mapping accuracy.

The paper analyzes whole genome sequences of 150,119 UK Biobank individuals, identifying variants and associations with traits, and provides a haplotype reference panel for imputation.

The paper was withdrawn due to premature posting and is not intended to be cited for the UK Biobank Pharma Proteomics Project.

The paper presents a modular and efficient workflow for single-cell RNA-seq pre-processing using kallisto and bustools, demonstrating its flexibility for RNA velocity analyses.

Bifrost constructs and indexes colored and compacted de Bruijn graphs efficiently for large genomic datasets, reducing memory usage and processing time.

The paper introduces the BUS format for single-cell RNA-seq pseudoalignments and the BUStools suite for analyzing data in this format.

The paper introduces a method to identify cell types and their transcriptional signatures from single-cell RNA-Seq data by analyzing transcript compatibility counts.

The paper compares genome alignment and transcriptome pseudoalignment, finding they produce similar quantifications and are interchangeable for many applications.

This study identifies mutations in RPL3L and MYZAP as risk factors for atrial fibrillation through genome-wide association analysis and functional validation.

Graphtyper efficiently genotypes sequence variants in large populations using pangenome graphs, achieving high accuracy and scalability.

Read pairs that fail kallisto pseudoalignment because of conflicting matches are filtered by pizzly to call, assemble and quantify gene fusions from raw RNA-seq in minutes rather than hours.

The paper introduces a method for differential RNA-Seq analysis that separates biological and inferential variance using bootstrapping and response error modeling, implemented in the sleuth app.

The paper presents a probabilistic method for RNA-seq quantification that achieves near-optimal accuracy.

chopBAI reduces I/O bottlenecks by splitting BAM index files, enabling efficient analysis of large sequencing cohorts.

BamHash verifies that FASTQ read pairs match BAM files, allowing safe deletion of raw data after alignment.

KmerStream estimates sequencing error rates and genome size using k-mer frequency statistics, showing variability in error rates between sequencing runs.