Hannah K. Wayment-Steele

Makeshift provides a lightweight Python package for accessing and analyzing NMR data, enabling reproducible and scalable analysis of biomolecular dynamics and chemical environments.

Trained on missing NMR amide assignments in the Biological Magnetic Resonance Data Bank, Dyna-1 (ESM-3 layer 22 embeddings) predicts micro-to-millisecond conformational exchange seen in relaxation data.

This study investigates whether genomic language models have learned RNA thermodynamics by analyzing their pairwise dependencies using the categorical Jacobian operation and comparing them to biophysical models.

The paper designs compact RNA sensors for complex functions of multiple inputs using crowdsourced designs and an algorithm, achieving applications like tuberculosis diagnosis.

SLAE learns all-atom protein representations from local atomic neighborhoods, achieving state-of-the-art performance in downstream tasks through transfer learning.

This paper refutes claims that AF-Cluster fails to use local evolutionary couplings in its MSA clusters and shows they play an important role in predictions.

This paper introduces a high-throughput method to measure DNA melting and develops models to improve DNA folding thermodynamics predictions.

This paper characterizes the conformational landscape of KaiB, revealing its slow interconversion between states and how evolution tunes this process to match circadian rhythms.

The paper investigates how ESM-2, a protein language model, predicts protein contacts by analyzing coevolving residue statistics and local sequence context.

This study develops PERSIST-seq and In-line-seq to optimize mRNA stability and translation, finding that in-cell stability is more critical than ribosome load for protein output.

The paper updates the Eterna100 RNA design benchmark to work with ViennaRNA 2 and shows that design performance depends on the folding engine used.

The paper identifies conserved RNA regions in SARS-CoV-2 and SARS-related viruses, predicts their secondary structures, and highlights potential antiviral and diagnostic targets.

A crowdsourced RNA design approach discovers efficient, reversible molecular sensors that couple small molecule inputs to protein binding and fluorogenic outputs.