Martin A. Smith

The paper models HLA diversity and viral evolution to assess T cell escape risk across populations, finding that escape mutations gain fitness with additional mutations.

ECSFinder improves prediction of evolutionarily conserved RNA structures by combining machine learning with thermodynamic and covariation features from existing tools.

The paper benchmarks RNA isoform reconstruction algorithms using nanopore sequencing data to evaluate their accuracy in transcriptome assembly.

This study demonstrates real-time transcriptomic profiling using nanopore sequencing and a neural network classifier to classify childhood acute lymphoblastic leukemia subtypes efficiently.

The paper differentiates iPSCs from EA/TEF patients and healthy donors into esophageal organoids, identifying SOX2 and NKX2.1 dysregulation as potential causes of foregut compartmentalization defects.

The paper presents computational tools for analyzing SARS-CoV-2 genetic data, showing that haplotype networks and unsupervised learning can reveal lineage emergence and population dynamics.

This study shows DNA hypomethylation disrupts replication timing precision and 3D genome organisation in colorectal cancer DNMT1 and DNMT3B knockout cells.

The paper introduces SLOW5, a new file format for nanopore sequencing data that accelerates analysis on HPC systems by enabling efficient parallelization.

This paper introduces a deep learning method for barcoding and demultiplexing direct RNA sequencing reads without base-calling or additional library steps, achieving high specificity and read demultiplexing rates.

This paper introduces f5c, a GPU-optimized implementation of Adaptive Banded Event Alignment for faster nanopore signal analysis and DNA methylation detection.

This paper demonstrates the detection of m6A RNA modifications in native RNA sequences using Oxford Nanopore Technologies with high accuracy.

SquiggleKit simplifies nanopore signal data manipulation by providing tools for file handling, visualization, and signal processing.

This paper introduces RAGE-seq, a method combining long-read and short-read sequencing to analyze lymphocyte clonal and transcriptional features in breast cancer tissues.

The paper identifies novel non-coding RNAs in human brain tissue associated with neuropsychiatric traits using targeted RNA capture and long-read sequencing.

The paper reduces memory requirements for long read alignment on mobile devices by optimizing parameters and merging techniques, enabling human genome alignment with 2GB RAM without accuracy loss.

The paper demonstrates that chiral DNA sequence pairs perform equivalently in clinical genomics techniques and proposes them as commutable reference standards to improve mutation detection in tumor samples.