Recent Advances in Machine Learning Variant Effect Prediction Tools for Protein Engineering.

Proteins are Nature's molecular machinery and comprise diverse roles while consisting of chemically similar building blocks. In recent years, protein engineering and design have become important research areas, with many applications in the pharmaceutical, energy, and biocatalysis fields, among others-where the aim is to ultimately create a protein given desired structural and functional properties. It is often critical to model the relationship between a protein's sequence, folded structure, and biological function to assist in such protein engineering pursuits. However, significant challenges remain in concretely mapping an amino acid sequence to specific protein properties and biological activities. Mutations may enhance or diminish molecular protein function, and the epistatic interactions between mutations result in an inherently complex mapping between genetic modifications and protein function. Therefore, estimating the quantitative effects of mutations on protein function(s) remains a grand challenge of biology, bioinformatics, and many related fields and would rapidly accelerate protein engineering tasks when successful. Such estimation is often known as variant effect prediction (VEP). However, progress has been demonstrated in recent years with the development of machine learning (ML) methods in modeling the relationship between mutations and protein function. In this Review, recent advances in variant effect prediction (VEP) are discussed as tools for protein engineering, focusing on techniques incorporating gains from the broader ML community and challenges in estimating biomolecular functional differences. Primary developments highlighted include convolutional neural networks, graph neural networks, and natural language embeddings for protein sequences.

Sodium ion channels as potential therapeutic targets for cancer metastasis.

Is it possible to develop drugs for the treatment of a specific type of metastatic cancer by targeting sodium ion channels?

Identification of TrkB Binders from Complex Matrices Using a Magnetic Drug Screening Nanoplatform.

Brain-derived neurotrophic factor (BDNF) and its receptor tyrosine receptor kinase B (TrkB) have been shown to play an important role in numerous neurological disorders, such as Alzheimer's disease. The identification of biologically active compounds interacting with TrkB serves as a drug discovery strategy to identify drug leads for neurological disorders. Here, we report effective immobilization of functional TrkB on magnetic iron oxide nanoclusters, where TrkB receptors behave as "smart baits" to bind compounds from mixtures and magnetic nanoclusters enable rapid isolation through magnetic separation. The presence of the immobilized TrkB was confirmed by specific antibody labeling. Subsequently, the activity of the TrkB on iron oxide nanoclusters was evaluated with ATP/ADP conversion experiments using a known TrkB agonist. The immobilized TrkB receptors can effectively identify binders from mixtures containing known binders, synthetic small molecule mixtures, and Gotu Kola (Centella asiatica) plant extracts. The identified compounds were analyzed by an ultrahigh-performance liquid chromatography system coupled with a quadrupole time-of-flight mass spectrometer. Importantly, some of the identified TrkB binders from Gotu Kola plant extracts matched with compounds previously linked to neuroprotective effects observed for a Gotu Kola extract approved for use in a clinical trial. Our studies suggest that the possible therapeutic effects of the Gotu Kola plant extract in dementia treatment, at least partially, might be associated with compounds interacting with TrkB. The unique feature of this approach is its ability to fast screen potential drug leads using less explored transmembrane targets. This platform works as a drug-screening funnel at early stages of the drug discovery pipeline. Therefore, our approach will not only greatly benefit drug discovery processes using transmembrane proteins as targets but also allow for evaluation and validation of cellular pathways targeted by drug leads.

Caffeine and Theophylline Inhibit ß-Galactosidase Activity and Reduce Expression in Escherichia coli.

The ß-galactosidase enzyme is a common reporter enzyme that has been used extensively in microbiological and synthetic biology research. Here, we demonstrate that caffeine and theophylline, common natural methylxanthine products found in many foods and pharmaceuticals, negatively impact both the expression and activity of ß-galactosidase in Escherichia coli. The ß-galactosidase activity in E. coli grown with increasing concentrations of caffeine and theophylline was reduced over sixfold in a dose-dependent manner. We also observed decreasing lacZ mRNA transcript levels with increasing methylxanthine concentrations in the growth media. Similarly, caffeine and theophylline inhibit the activity of the purified ß-galactosidase enzyme, with an approximately 1.7-fold increase in K M toward o-nitrophenyl-ß-galactoside and a concomitant decrease in v max. The authors recommend the use of alternative reporter systems when such methylxanthines are expected to be present.