Metabolism underpins cellular function by supplying energy, biosynthetic precursors, and redox balance and in yeast there are thousands of metabolic reactions that are tightly coordinated through multilayered regulation. The yeast Saccharomyces cerevisiae has become a central model for studying metabolism and its regulation and following publication of its genome in 1996, this yeast became pivotal in systems biology. Systems biology integrates experimental data with mathematical modeling to analyze complex cellular networks. A major advance for metabolic analysis was the development of flux balance analysis and genome sequencing enabled reconstruction of the first genome-scale metabolic model (GEM) for yeast. This initial GEM described how hundreds of genes, reactions, and metabolites interact across compartments. Subsequent models, including Yeast8 and Yeast9, expanded the coverage and predictive power, and these models enable metabolic comparison, physiological analysis, omics integration, and design of strains that can be used for production of chemicals and biopharmaceuticals. Overall, S. cerevisiae remains a cornerstone of systems biology and biotechnology, with continued advances expected in integrative modeling and engineering applications.

RNA alternative splicing is a fundamental post-transcriptional mechanism whose dysregulation drives various human diseases. Predicting splicing outcomes is therefore a central challenge in precision medicine. This Review traces the evolution of computational approaches from early statistical heuristics to modern artificial intelligence frameworks. We dissect the methodologies that shape predictive performance, including training data scale, output resolution, splicing event quantification and model complexity. Among these factors, quantitative assessment of splicing events and the increasing complexity of model architectures represent particularly critical axes that define both biological interpretability and computational feasibility. We further describe how these models empower translational applications, from annotating variant effects to guiding antisense oligonucleotide development. Nonetheless, persistent challenges remain, including the interpretation of deep-intronic mutations, isoform-level reconstruction and integration of multimodal data. Together, these perspectives define both the progress achieved and the opportunities ahead for splicing prediction in genomics and medicine. Predicting splicing outcomes is a central challenge in precision medicine. This Review traces the evolution of computational approaches from early statistical heuristics to modern artificial intelligence frameworks to characterize splicing events.

Bitter taste is a critical quality determinant in food systems, particularly those using sustainable protein hydrolysates, where the unpredictable formation of bitter peptides severely limits consumer acceptance. Achieving predictive control over flavor chemistry requires deciphering the complex sequence-activity relationship. To address this, we integrated the generative capacity of a protein language model with BitterPep-GCN, a Graph Convolutional Network (GCN) capable of robust in silico bitter/non-bitter classification, to target the de novo design of functional bitter and non-bitter sequences. We achieved this by generating two strategic peptide libraries: a targeted tripeptide library derived from known bitter and non-bitter peptide sequences, and a set of de novo designed sequences. For the de novo designed peptides, we fine-tuned the conditional language model ZymCTRL on our curated dataset of sensory-validated bitter peptides (BPS-1000). Both libraries were subjected to classification and rigorous filtering using BitterPep-GCN to select high-confidence candidates for validation. The selected peptides were purchased and rigorously assessed for high purity. Sensory tests were conducted by an expert human panel to determine intrinsic taste quality and taste recognition thresholds. The results validated the high predictive fidelity of our pipeline: out of the 31 tested peptides, 25 were correctly classified, including 15 confirmed bitter and 10 confirmed non-bitter sequences. This study successfully demonstrates the application of machine learning frameworks in the design of bioactive peptides. It provides a set of novel taste-active peptides that can be used to accelerate the rational mitigation of off-tastes in next-generation food products.

Plants and pathogens engage in complex biochemical communication, mediated by proteins, specialized metabolites, phytohormones and phytohormone-mimicking compounds. These interactions drive a dynamic ‘co-evolutionary arms race’, as plants and microbes compete to gain an advantage, ultimately transforming the infection site into a distinct micro-ecosystem. Microbe-induced galls are abnormal plant organogenesis induced by specific pathogens, creating a niche where the pathogen manipulates the host plant machinery to ensure its own survival. Such galls adversely affect agricultural and horticultural productivity by stunting plant growth and causing deformities. However, the rapid cell proliferation in tumourigenesis, along with the mechanisms driving robust shoot and root re-differentiation, present exciting opportunities for innovative biotechnological applications. Therefore, elucidation of these mechanisms is crucial for advancing basic research with significant potential for agricultural applications. This review focuses on galls induced by Agrobacterium tumefaciens and Rhodococcus fascians—two phytopathogens that utilize phytohormones as tumor-inducing molecules—to highlight the mechanisms underlying plant–pathogen interactions within this specialized microenvironment. It also explores the evolutionary adaptations and strategies of these pathogens. Gaining insight into these biological processes is key to understanding the mechanisms driving biological diversity and evolution, with implications extending beyond plant pathology into the broader field of molecular plant physiology.

The gut microbiota is implicated in adverse effects associated with low-calorie sweeteners. Yet, the direct impact of sweeteners on gut bacteria remains largely uncharacterized. Here, we report interactions between 25 phylogenetically diverse gut bacterial strains and 39 commercially used sweeteners. We tested these sweeteners individually and in combination with four commonly co-consumed compounds, viz., advantame, caffeine, vanillin, and duloxetine. Three-quarters of the tested sweeteners individually impacted the growth of at least one tested bacterial strain. Further, over 100 interactions were found between sweeteners and the four co-consumed compounds. Isosteviol, a commonly used sweetener-component, and duloxetine, an antidepressant, synergistically inhibited Roseburia intestinalis, a bacterium previously linked to glucose homeostasis, and Parabacteroides merdae, a prevalent commensal linked to healthy microbiota. Proteomic, metabolomic, and genetic analyses indicate altered small molecule transport underpinning this sweetener–drug synergy. The isosteviol-duloxetine combination also modulated metabolism of a synthetic gut bacterial community, leading to increased toxicity to HeLa cells and altered secretion of inflammation-modulatory cytokines IL-6 and IL-8 by Caco-2 cells. Our data warrant further studies on interactions between low-calorie sweeteners and common xenobiotics.

CRISPR-Cas and toxin-antitoxin systems can serve as antiviral defense mechanisms in prokaryotes. In typical toxin-antitoxin systems, toxin activation can limit phage propagation by inducing growth arrest or reduced cellular fitness, while the antitoxin neutralizes toxin activity. Here, we study potential functional synergy between a CRISPR-Cas13a system and a type II toxin-antitoxin module (HicAB) from a Leptotrichia bacterium, when heterologously expressed in E. coli, as well as in biochemical and structural analyses. We show that the antitoxin HicB exhibits toxic properties, and Cas13a directly activates HicB, triggering growth inhibition and conferring protection against bacteriophages. Structural analyses reveal that Cas13a binding promotes the spatial proximity of HicB tetramers, likely enabling its activation. The toxin HicA competitively binds to HicB, thereby inhibiting Cas13a-mediated HicB activation. Importantly, both CRISPR RNA and HicB independently suppress HicA toxicity. Structural evidence indicates that CRISPR RNA forms a hetero-tetradecameric complex with HicAB, occluding HicA’s active site and neutralizing its toxic function. Thus, our findings indicate functional synergy between distinct bacterial immune strategies. CRISPR-Cas and toxin-antitoxin systems can serve as antiviral defense mechanisms in prokaryotes. Here, the authors provide evidence of functional synergy between a CRISPR-Cas13a system and a type II toxin-antitoxin module.

Polyethylene terephthalate (PET) hydrolases have been extensively studied for their potential applications in plastic degradation. However, the structural and mechanistic factors that limit their catalytic efficiency are not yet fully understood. Here, we identify the protruding, surface-exposed C-terminal loop (SEC-loop) in Cryptosporangium aurantiacum PETase (CaPETase) that negatively impacts enzymatic activity by restricting productive access of enzyme to PET. Loop replacement experiments show the non-protruding SEC-loop enhances PET depolymerization rates, despite being ~25 Å from the active site. Kinetic and adsorption studies indicate the non-protruding SEC-loop promotes productive PET access to the enzyme without affecting binding affinity. To further assess the broader applicability of this strategy across diverse PETases, SEC-loop replaced variants of representative PETases are characterized through kinetic and adsorption analyses. We show an engineering strategy focused on modulating enzyme accessibility rather than simply modifying the catalytic site, in rational enzyme design aimed at improving PET degradation efficiency. Polyethylene terephthalate (PET) hydrolases have been extensively studied for their applications in plastic degradation, but the structural and mechanistic factors that limit their catalytic efficiency are not yet fully understood. Here, the authors identify the protruding, surface-exposed C-terminal loop (SEC-loop) in Cryptosporangium aurantiacum PETase that negatively impacts enzymatic activity by restricting productive access of enzyme to PET substrates.

Methanol is a promising C1 feedstock for bioproduction but its cytotoxicity limits applications. Here we enhanced Komagataella phaffii GS115 methanol tolerance to 10–13% (v/v) by overexpressing four genes (phosphatidylethanolamine methyltransferase (PET2), phosphatidylserine synthase (PSA), PAS_chr1-4_0178 (0178) and PAS_chr2-2_0267 (0267)) (S33 strain). Our results show that, under high methanol stress, S33 undergoes membrane lipid remodelling-driven metabolic reprogramming, with central carbon flux being redistributed from the pentose phosphate pathway and citric acid cycle towards glycolysis. Within this framework, 0178 encodes a vacuolar transporter of sulfur-containing amino acids, which we propose supports phospholipid methylation and redox balance. Meanwhile, 0267 encodes a peroxisomal thiolase-degrading long‑chain fatty acyl‑CoAs, limiting toxic long-chain sphingolipid accumulation, maintaining membrane stability. These genes enable S33 to achieve lipid titre of approximately 40 g l−1 under non-sterile fed‑batch fermentation conditions. This work positions S33 as a methanol-tolerant K. phaffii chassis for further strain and process development. Bioproduction from methanol is often challenging owing to the cytotoxicity of methanol. Here a methanol-tolerant Komagataella phaffii chassis is engineered by overexpressing four membrane genes, increasing tolerance to 10–13% methanol and enabling non-sterile lipid production (>40 g l−1) via membrane lipid remodelling and metabolic rewiring.