C. O. Wilke

Semantic search using protein large language models detects class II microcins in bacterial genomes

The paper uses protein large language models to detect class II microcins in bacterial genomes, finding embeddings outperform sequence-based methods like BLAST in identifying known and novel microcins.

A computational model for bacteriophage {varphi}X174 gene expression

This study constructs a computational model of bacteriophage φX174 gene regulation and shows that canonical promoters and terminators suffice to explain transcript abundance variation.

Deep mutational scanning and machine learning uncover antimicrobial peptide features driving membrane selectivity

The paper uses deep mutational scanning and machine learning to identify sequence features in antimicrobial peptides that enhance bacterial membrane selectivity while reducing mammalian toxicity.

Two sequence- and two structure-based ML models have learned different aspects of protein biochemistry

The paper compares sequence- and structure-based ML models for predicting protein residue properties, finding they excel at different tasks and that combining them improves accuracy.

Growth-dependent gene expression variation influences the strength of codon usage biases

The paper shows that growth rate-dependent gene expression variation significantly influences codon usage biases in microbial genomes.

Recombination rate inference via deep learning is limited by sequence diversity

This study shows that deep learning models for recombination rate inference perform poorly on low-diversity sequence data, highlighting inherent information limits in such datasets.

Designing β-hairpin peptide macrocycles for antibiotic potential

The paper designs and screens a synthetic macrocyclic β-hairpin peptide library for antibiotic potential, identifying active peptides with unique structural and charge features.

Site-specific amino acid distributions follow a universal shape

The paper shows that a single-parameter model can accurately predict amino acid frequency ranks across protein sequences, revealing a universal distribution shape.

A comprehensive co-expression network analysis in Vibrio cholerae

This paper constructs a gene co-expression network for Vibrio cholerae using RNA-sequencing data to identify novel gene interactions and improve functional annotations.

Moderate amounts of epistasis are not evolutionarily stable in small populations

The paper shows that moderate epistasis is not evolutionarily stable in small populations, as high mutation rates lead to shifts toward either lower or higher epistasis depending on initial conditions.

Phylogenetic weighting does little to improve the accuracy of evolutionary coupling analyses

This study evaluates various weighting methods for evolutionary coupling analysis and finds that current best practices outperform alternatives, with minor benefits from sequence weighting compared to post-hoc scaling.

Evolutionary couplings detect side-chain interactions

This paper shows evolutionary coupling algorithms better detect side-chain interactions than backbone interactions, highlighting the importance of considering interaction types in contact prediction.

The many nuanced evolutionary consequences of duplicated genes

This study investigates how gene duplication affects the evolution of interacting proteins, finding that duplicated genes can drive unexpected functional changes in their partners through dosage imbalance.

Combinatorial Approaches to Viral Attenuation

This paper develops a method for designing attenuated viruses using promoter ablation in bacteriophage T7 and shows predictable attenuation effects.

Selection removes Shine-Dalgarno-like sequences from within protein coding genes

The paper shows that Shine-Dalgarno sequences within bacterial genes are less conserved and likely deleterious, leading to reduced ATG start codons downstream.

Beyond thermodynamic constraints: Evolutionary sampling generates realistic protein sequence variation

This study shows that evolutionary simulation generates more realistic protein sequence variation than current design methods, highlighting the need for realistic sampling schemes.

Accelerated simulation of evolutionary trajectories in origin-fixation models

This paper introduces an accelerated algorithm for simulating origin-fixation models in evolutionary biology, reducing computational cost while maintaining accuracy in key evolutionary metrics.