Martin J. Lercher

The paper models growth costs of suboptimal protein allocation in microbial cells using growth balance analysis, showing how protein levels affect growth rates and fitness.

The paper uses single-cell RNA-sequencing to analyze cell diversity in Arabidopsis seeds, identifying distinct cell clusters and spatial domains along developmental axes.

The paper analyzes bacterial genomes to show that horizontally acquired genes are mostly lost quickly but some persist long-term, often involved in transport and metabolism.

The paper shows that high accuracy in predicting enzyme kcat fold-changes from sequence data often results from data biases, not true mechanistic understanding, and introduces a bias-aware framework to improve predictions.

FoldVision is a compute-efficient 3D CNN for proteins that outperforms sequence-based models on structure-dependent tasks using atom-level representations.

This study compares HGT inference methods, finding that gene presence-absence analysis outperforms gene tree-based approaches in detecting horizontal gene transfer.

This study introduces Growth Mechanics to show that metabolic oscillations in growing cells are necessary for maximizing fitness by optimizing resource allocation.

The paper finds that bacteria entering new environments have smaller genomes and lower horizontal gene transfer rates, challenging the assumption that genomic plasticity drives adaptation.

This paper evaluates the DLKcat model's ability to predict kcat values for enzymes and mutants, finding it unreliable for low sequence identity and poor at capturing mutation effects.

This paper introduces SPOT, a machine learning model that predicts substrates for transport proteins with high accuracy, outperforming existing methods.

The paper shows that 49% of bacterial gene families are preferentially located near the origin or terminus of replication, supporting growth rate-dependent expression optimization.

The paper analyzes proteome efficiency across metabolic pathways in E. coli, finding that efficiency increases along the nutrient flow, with costly pathways near optimal and transporters and central metabolism over-abundant.

ProSmith predicts drug-target interactions using a multi-modal Transformer Network, showing better generalization and performance than previous models.

This paper introduces TurNuP, a model predicting enzyme turnover numbers for natural reactions, outperforming existing methods and generalizing across enzyme similarity.

The paper develops two simplified growth balance analysis models for Synechocystis sp. PCC 6803 to predict cellular resource allocation under different environmental conditions and photodamage effects.

This paper introduces ESP, a machine learning model predicting enzyme-substrate pairs with over 90% accuracy, outperforming existing models for specific enzyme families.

The paper derives an optimal growth law for RNA composition in bacteria and shows that tRNA/rRNA ratio declines with growth rate, partially regulated by gene location.

The paper develops a deep learning model to predict Michaelis constants for enzyme-substrate combinations using structural features.

This paper develops numerical methods to optimize the growth rate of genome-scale metabolic models by treating enzyme concentrations as variables and clarifying simplifying assumptions in the optimization framework.

This study shows that phages and plasmids increase prokaryotic genome size, while CRISPR reduces it, with CRISPR targeting phages more than plasmids.

The paper shows that alleles with reduced enzyme function in metabolic genes cause increasing pleiotropy, often through effects on currency metabolites like ATP and NADPH.

The paper models resource allocation in C3 and C4 plants to show that C4 plants have limited phenotypic plasticity, suggesting their resource allocation reflects ancestral environments rather than current conditions.

This paper automates the reconstruction of high-quality genome-scale metabolic models using high-throughput data, demonstrating accuracy comparable to manual methods for Streptococcus strains.

The paper shows that most metabolic adaptations in E. coli arise from single DNA segment acquisitions, not neutral processes, and highlights the power of horizontal gene transfer in enabling rapid environmental adaptation.

The paper derives a model predicting metabolite concentrations from enzyme concentrations and substrate affinities, showing high accuracy across E. coli and eukaryotes.