arXiv subjects

Mehdi D. Davari

Publications and source records attributed to Mehdi D. Davari.

6 recordsLinked to original sources

arXiv2026

Multitask Bayesian Neural Networks for Multiparameter Protein Engineering

Simultaneously engineering multiple protein properties remains a major challenge. Existing machine learning-based pipelines for protein engineering often model properties separately, failing to capture their dependencies and trade-offs. Here, we systematically evaluate how Bayesian parameterization on Multitask Neural Networks can enable robust simultaneous protein engineering under scarce, noisy experimental data. We curated a comprehensive set of 27 multiparameter protein datasets. Then, we compared three algorithm architectures spanning low to full Bayesian parameterization across 16 sequence representations and dimensionality reduction (2,592 models). Bayesian Last Layer models delivered the strongest overall accuracy, generalization, and calibration, ranking as the top-performing model on 70% of benchmark datasets. Dimensionality reduction improved predictive performance by up to 42% and enhanced calibration up to 57% across architectures. Notably, simple One-Hot encoding achieved top performance on 25% of benchmark datasets, particularly with larger datasets. These results establish practical design principles for reliable and data-efficient multiparameter protein engineering.

q-bio.BM↗

arXiv2025

Decoding Polyphenol-Protein Interactions with Deep Learning: From Molecular Mechanisms to Food Applications

Polyphenols and proteins are essential biomolecules that influence food functionality and, by extension, human health. Their interactions -- hereafter referred to as PhPIs (polyphenol-protein interactions) -- affect key processes such as nutrient bioavailability, antioxidant activity, and therapeutic efficacy. However, these interactions remain challenging due to the structural diversity of polyphenols and the dynamic nature of protein binding. Traditional experimental techniques like nuclear magnetic resonance (NMR) and mass spectrometry (MS), along with computational tools such as molecular docking and molecular dynamics (MD), have offered important insights but face constraints in scalability, throughput, and reproducibility. This review explores how deep learning (DL) is reshaping the study of PhPIs by enabling efficient prediction of binding sites, interaction affinities, and MD using high-dimensional bio- and chem-informatics data. While DL enhances prediction accuracy and reduces experimental redundancy, its effectiveness remains limited by data availability, quality, and representativeness, particularly in the context of natural products. We critically assess current DL frameworks for PhPIs analysis and outline future directions, including multimodal data integration, improved model generalizability, and development of domain-specific benchmark datasets. This synthesis offers guidance for researchers aiming to apply DL in unraveling structure-function relationships of polyphenols, accelerating discovery in nutritional science and therapeutic development.

q-bio.BM↗

arXiv2025

Best Practices for Machine Learning-Assisted Protein Engineering

Data-driven modeling based on Machine Learning (ML) is becoming a central component of protein engineering workflows. This perspective presents the elements necessary to develop effective, reliable, and reproducible ML models, and a set of guidelines for ML developments for protein engineering. This includes a critical discussion of software engineering good practices for development and evaluation of ML-based protein engineering projects, emphasizing supervised learning. These guidelines cover all the necessary steps for ML development, from data acquisition to model deployment. Additionally, the present perspective provides practical resources for the implementation of the outlined guidelines. These recommendations are also intended to support editors and scientific journals in enforcing good practices in ML-based protein engineering publications, promoting high standards across the community. With this, the aim is to further contribute to improved ML transparency and credibility by easing the adoption of software engineering best practices into ML development for protein engineering. We envision that the wide adoption and continuous update of best practices will encourage informed use of ML on real-world problems related to protein engineering.

q-bio.BM↗

arXiv2025

Machine Learning-Driven Enzyme Mining: Opportunities, Challenges, and Future Perspectives

Enzyme mining is rapidly evolving as a data-driven strategy to identify biocatalysts with tailored functions from the vast landscape of uncharacterized proteins. The integration of machine learning into these workflows enables high-throughput prediction of enzyme functions, including Enzyme Commission numbers, Gene Ontology terms, substrate specificity, and key catalytic properties such as kinetic parameters, optimal temperature, pH, solubility, and thermophilicity. This review provides a systematic overview of state-of-the-art machine learning models and highlights representative case studies that demonstrate their effectiveness in accelerating enzyme discovery. Despite notable progress, current approaches remain limited by data scarcity, model generalizability, and interpretability. We discuss emerging strategies to overcome these challenges, including multi-task learning, integration of multi-modal data, and explainable AI. Together, these developments establish ML-guided enzyme mining as a scalable and predictive framework for uncovering novel biocatalysts, with broad applications in biocatalysis, biotechnology, and synthetic biology.

q-bio.BM↗

arXiv2025

Geometric deep learning assists protein engineering. Opportunities and Challenges

Protein engineering is experiencing a paradigmatic shift through the integration of geometric deep learning into computational design workflows. While traditional strategies, such as rational design and directed evolution, have enabled relevant advances, they remain limited by the complexity of sequence space and the cost of experimental validation. Geometric deep learning addresses these limitations by operating on non-Euclidean domains, capturing spatial, topological, and physicochemical features essential to protein function. This perspective outlines the current applications of GDL across stability prediction, functional annotation, molecular interaction modeling, and de novo protein design. We highlight recent methodological advances in model generalization, interpretability, and robustness, particularly under data-scarce conditions. A unified framework is proposed that integrates GDL with explainable AI and structure-based validation to support transparent, autonomous design. As GDL converges with generative modeling and high-throughput experimentation, it is emerging as a central technology in next-generation protein engineering and synthetic biology.

q-bio.QM↗

arXiv2025

From thermodynamics to protein design: Diffusion models for biomolecule generation towards autonomous protein engineering

Protein design with desirable properties has been a significant challenge for many decades. Generative artificial intelligence is a promising approach and has achieved great success in various protein generation tasks. Notably, diffusion models stand out for their robust mathematical foundations and impressive generative capabilities, offering unique advantages in certain applications such as protein design. In this review, we first give the definition and characteristics of diffusion models and then focus on two strategies: Denoising Diffusion Probabilistic Models and Score-based Generative Models, where DDPM is the discrete form of SGM. Furthermore, we discuss their applications in protein design, peptide generation, drug discovery, and protein-ligand interaction. Finally, we outline the future perspectives of diffusion models to advance autonomous protein design and engineering. The E(3) group consists of all rotations, reflections, and translations in three-dimensions. The equivariance on the E(3) group can keep the physical stability of the frame of each amino acid as much as possible, and we reflect on how to keep the diffusion model E(3) equivariant for protein generation.

q-bio.QM↗