Stanislav Mazurenko

The paper benchmarks variant effect predictors using a leakage-controlled dataset and develops a gradient-boosting consensus model that outperforms existing tools.

Unattended droplet-microfluidic Bayesian optimization of a glycerol detection cascade matched design of experiments (DoE) with one third of the conditions; a second cascade used DoE with Raman readout.

PRIZM uses low-N data and zero-shot models to rank protein variants, improving design efficiency with minimal experimental data.

This study uses ultrahigh-throughput screening to identify mutations in dehalogenases that modulate substrate access and binding, revealing distal functional effects on enzyme dynamics.

This study uses zero-shot models to predict activity-enhancing mutations in fluorinase, leading to improved catalytic efficiency and stability in engineered variants.

The paper compares biochemical and immunological properties of engineered low-immunogenic staphylokinase variants to identify the most promising for thrombolytic therapy.

ProtXAI combines molecular dynamics and explainable AI to identify residue-level determinants of protein conformational changes across diverse systems.

The paper evaluates deep learning models for predicting protein melting temperatures using proteomics and biophysics datasets, finding that fine-tuned embeddings improve performance on diverse validation sets.

This study shows taurine inhibits ApoE4 aggregation and reduces its pathophysiological effects, similar to tramiprosate and 3-sulfopropanoic acid.

This study investigates the conformational flexibility of staphylokinase variants to explain their reduced immunogenicity and altered properties.

The paper introduces SoluProtMut, a Siamese deep learning model for predicting solubility effects of protein mutations, validated experimentally on 277 mutants.

The paper introduces a computational method to predict protein evolution by reconstructing a protein's evolutionary history and suggesting future amino acid substitutions to improve desired properties.

PredictONCO is a web tool that predicts the impact of mutations in cancer-related proteins using structural and sequence-based features, achieving high accuracy in variant classification.

The paper develops a computational pipeline to annotate functional tunnels in enzyme-ligand complexes and validates it using molecular systems, showing it matches molecular dynamics simulations in speed and accuracy.

This paper presents a bioinformatics pipeline for analyzing missense mutations in oncogenic proteins, providing a probability score for functional impact and suggesting potential FDA/EMA-approved drugs.