The quest to predict and understand protein evolution has been hindered by limitations on both the theoretical and the experimental fronts. Most existing theoretical models of evolution are descriptive, rather than predictive, leaving the final modifications in the hands of researchers. Existing experimental techniques to help probe the evolutionary sequence space of proteins, such as directed evolution, are resource-intensive and require specialised skills. We present the successor sequence predictor (SSP) as an innovative solution. Successor sequence predictor is an in silico protein design method that mimics laboratory-based protein evolution by reconstructing a protein's evolutionary history and suggesting future amino acid substitutions based on trends observed in that history through carefully selected physicochemical descriptors. This approach enhances specialised proteins by predicting mutations that improve desired properties, such as thermostability, activity, and solubility. Successor Sequence Predictor can thus be used as a general protein engineering tool to develop practically useful proteins. The code of the Successor Sequence Predictor is provided at https://github.com/loschmidt/successor-sequence-predictor , and the design of mutations will be also possible via an easy-to-use web server https://loschmidt.chemi.muni.cz/fireprotasr/ . SCIENTIFIC CONTRIBUTION: The Successor Sequence Predictor advances protein evolution prediction at the amino acid level by integrating ancestral sequence reconstruction with a novel in silico approach that models evolutionary trends through selected physicochemical descriptors. Unlike prior work, SSP can forecast future amino acid substitutions that enhance protein properties such as thermostability, activity, and solubility. This method reduces reliance on resource-intensive directed evolution techniques while providing a generalizable, predictive tool for protein engineering.

• Klíčová slova
• Activity, Adaptation, Evolution, Evolutionary trajectory, Protein design, Solubility, Thermostability,
• Publikační typ
• časopisecké články MeSH

Recombinant proteins play pivotal roles in numerous applications including industrial biocatalysts or therapeutics. Despite the recent progress in computational protein structure prediction, protein solubility and reduced aggregation propensity remain challenging attributes to design. Identification of aggregation-prone regions is essential for understanding misfolding diseases or designing efficient protein-based technologies, and as such has a great socio-economic impact. Here, we introduce AggreProt, a user-friendly webserver that automatically exploits an ensemble of deep neural networks to predict aggregation-prone regions (APRs) in protein sequences. Trained on experimentally evaluated hexapeptides, AggreProt compares to or outperforms state-of-the-art algorithms on two independent benchmark datasets. The server provides per-residue aggregation profiles along with information on solvent accessibility and transmembrane propensity within an intuitive interface with interactive sequence and structure viewers for comprehensive analysis. We demonstrate AggreProt efficacy in predicting differential aggregation behaviours in proteins on several use cases, which emphasize its potential for guiding protein engineering strategies towards decreased aggregation propensity and improved solubility. The webserver is freely available and accessible at https://loschmidt.chemi.muni.cz/aggreprot/.

• MeSH
• algoritmy MeSH
• internet * MeSH
• konformace proteinů MeSH
• neuronové sítě MeSH
• proteinové agregáty * MeSH
• proteinové inženýrství metody   MeSH
• proteiny chemie   genetika   MeSH
• rozpustnost MeSH
• sbalování proteinů MeSH
• software * MeSH
• Publikační typ
• časopisecké články MeSH
• Názvy látek
• proteinové agregáty * MeSH
• proteiny MeSH

Recent progress in engineering highly promising biocatalysts has increasingly involved machine learning methods. These methods leverage existing experimental and simulation data to aid in the discovery and annotation of promising enzymes, as well as in suggesting beneficial mutations for improving known targets. The field of machine learning for protein engineering is gathering steam, driven by recent success stories and notable progress in other areas. It already encompasses ambitious tasks such as understanding and predicting protein structure and function, catalytic efficiency, enantioselectivity, protein dynamics, stability, solubility, aggregation, and more. Nonetheless, the field is still evolving, with many challenges to overcome and questions to address. In this Perspective, we provide an overview of ongoing trends in this domain, highlight recent case studies, and examine the current limitations of machine learning-based methods. We emphasize the crucial importance of thorough experimental validation of emerging models before their use for rational protein design. We present our opinions on the fundamental problems and outline the potential directions for future research.

• Publikační typ
• časopisecké články MeSH
• přehledy MeSH