Kevin M. Esvelt

The paper proposes Societal Safety Levels (SSLs) to complement Biosafety Levels (BSLs) by setting quantitative thresholds for expected public fatalities from laboratory research on pandemic pathogens.

The paper develops a rubric to define sequences of concern by analyzing test sequences and refining categorization through stakeholder review, reducing disputed categorization by 44.3% for controlled viruses.

This paper demonstrates a gene drive strategy targeting KLHL10 in mice to increase homing efficiency by removing NHEJ-mutated sperm, but observes unexpected mutation absence in female offspring.

This study genetically engineers mice to produce a protective antibody against Lyme disease, demonstrating heritable immunity that interrupts pathogen transmission.

The paper develops and characterizes gene drive docking lines in Anopheles gambiae using RMCE and HACK techniques, identifies challenges for robust gene drive development.

The paper models the impact of virus discovery and pandemic virus identification on preventing future pandemic deaths, finding that virus discovery has greater potential than laboratory-based pandemic potential characterization.

FLIGHTED improves protein fitness model performance by accounting for experimental noise in high-throughput data, showing that data size and architecture matter more than model scale.

The paper introduces a method combining exact-match search with functional variant prediction to automate DNA synthesis screening for controlled genes.

The paper introduces virus-like DNA tracers to improve the accuracy and resolution of wastewater-based pathogen monitoring systems by assessing assay performance and signal decay.

This paper presents a framework to identify fertility gene targets for mammalian pest control by evaluating gene function, expression, and knockout results, identifying 16 male and 12 female fertility genes conserved across invasive species.

This paper develops a CRISPR-based genetic sterilization system for Anopheles gambiae males, achieving high infertility rates and population suppression in competition cages.

This study examines how E. coli's genetic code tolerates tRNA mutations that increase codon size, finding that while many tRNAs can be converted to quadruplet tRNAs, only nine aminoacyl tRNA synthetases tolerate quadruplet anticodons, limiting the chemical alphabet of a synthetic quadruplet code.

This paper introduces DHARMA, an ultra-high-throughput method for measuring biomolecular activities, and benchmarks protein models using large variant datasets.

This paper introduces a data-efficient deep learning approach for protein engineering that uses minimal experimental data to predict and optimize protein function with high throughput.

This paper develops a machine learning toolkit to attribute genetic engineering to labs by analyzing DNA motifs and phenotype, achieving 70% accuracy in distinguishing over 1,300 labs.

This study shows that CRISPR gene drives can invade wild populations even with resistance alleles, challenging claims that they are safe for release.

The paper introduces a daisy quorum drive system designed to alter wild populations within defined boundaries and restore engineered genes to their original state.

This paper describes a gene drive system using repeated genomic elements as a generational clock to limit the spread of engineered traits.