James B. Brown

Exploiting regulatory heterogeneity to systematically identify enhancers with high accuracy

The paper identifies a class of enhancers with high prediction accuracy by focusing on regulatory heterogeneity in Drosophila, achieving 86.7% precision and 98% recall.

Refining interaction search through signed iterative Random Forests

This paper introduces signed iRF (s-iRF), a method to refine interaction search in biological systems, identifying stable and predictive feature interactions for enhancer activity and spatial gene expression.

iterative Random Forests to discover predictive and stable high order interactions

The paper introduces iRF, an algorithm for detecting stable high-order genomic interactions, and demonstrates its application in enhancer activity and alternative splicing.

De novo Identification of DNA Modifications Enabled by Genome-Guided Nanopore Signal Processing

Base-by-base tests of raw nanopore current, native versus amplified DNA, detect 4mC, 5mC and 6mA without training data, recovering seven methylases' motifs (AUC 0.62–0.97).