Combinatorial mutagenesis is essential for exploring protein sequence-function landscapes in engineering applications. However, while large-scale machine learning benchmarks exist for protein function prediction, they are primarily limited to single-mutant libraries, leaving a critical gap for combinatorial mutagenesis. Here we introduce CombinGym, a benchmarking platform featuring 14 curated combinatorial mutagenesis datasets spanning 9 proteins with diverse functional properties including binding affinity, fluorescence, and enzymatic activities. We evaluated nine machine learning algorithms from five methodological categories (alignment-based, protein language, structure-based, sequence-label, and substitution-based) across multiple prediction tasks, assessing both zero-shot and supervised learning performance using Spearman's ρ and Normalized Discounted Cumulative Gain metrics. Our analysis reveals the substantial impact of measurement noise and data processing strategies on model performance. By implementing hierarchical dataset splits (0-vs-rest, 1-vs-rest, 2-vs-rest, and 3-vs-rest scenarios), we demonstrate the value of lower-order mutation data for empowering machine learning models to predict higher-order mutant properties. We validated this capacity through both in silico simulation (improving fluorescence brightness of an oxygen-independent fluorescent protein) and experimental validation (engineering enzyme substrate specificity), achieving a substantial increase in specific activity. All datasets, benchmarks, and metrics are available through an interactive website (https://www.combingym.org), facilitating collaborative dataset expansion and model development through integration with automated biofoundry platforms.

The rapid expansion of protein sequence databases continues to outpace functional characterization, limiting efficient enzyme discovery in large, heterogeneous, and sparsely annotated sequence spaces. Here, we present an embedding-guided framework for structured navigation of enzyme sequence space, implemented in SelectZyme. The workflow integrates protein language model embeddings with dimensionality reduction, hierarchical clustering, connectivity reconstruction, and quantitative dendrogram analysis to enable exploration without reliance on fixed sequence identity thresholds or predefined functional annotations. Across distinct case studies, we demonstrate that embedding-defined neighborhoods preserve fold-level coherence even when sequence identity resides in the twilight zone and that biologically meaningful functional organization can emerge under fully unsupervised conditions. In a large multi-family PETase landscape exceeding 100,000 sequences, the framework supports scalable, anchor-guided prioritization under sparse labeling and process-motivated constraints. By unifying latent representations with connectivity-aware and hierarchical interpretation, this approach reframes enzyme mining from threshold-driven similarity filtering into structured exploration, providing a scalable methodological foundation for hypothesisdriven biocatalyst discovery and reasonable starting points for downstream protein engineering.

Bioelectrocatalytic conversion of organic waste uses renewable electricity to convert organic waste into high-value chemicals and fuels under mild conditions. It is regarded as a strategic link connecting waste refinement and the green economy, providing a sustainable approach for the high-value utilization of waste resources. This article systematically reviews: (1) the basic principles and reaction mechanisms of bioelectrocatalytic conversion of organic waste; (2) the development history and key breakthroughs of organic waste biological electrochemical conversion; (3) microbial engineering strategies based on genetic engineering and synthetic biology; (4) the directions for process and reactor optimization. This review uniquely integrates advancements in biological innovation and process engineering, offering a holistic perspective on synergistic optimization across molecular, microbial, and reactor scales. By explicitly bridging microbial carbon/electron flux engineering with bioreactor design and process coupling, it aims to establish a comprehensive roadmap for accelerating the industrial-scale application of bioelectrocatalytic technologies in organic waste treatment.

Messenger RNA (mRNA) has rapidly emerged as a transformative therapeutic modality, exemplified by its growing applications in infectious diseases, oncology, and genetic disorders. The chemical programmability of mRNA allows researchers to modulate its function by introducing synthetic modifications across the molecule─from the cap structure, untranslated regions (UTRs), coding sequence (CDS) to poly(A) tail and from base, backbone to ribose sugars. Beyond sequence-level design, recent advances have introduced a new dimension of control: topological engineering. Circular RNAs, branched structures, and synthetic lariat architectures are reshaping how we approach RNA stability, immunogenicity, and translation. This review surveys recent advances in the chemical and topological engineering of mRNA, emphasizing four key areas: (1) enzymatic, chemical, and hybrid methodologies that expand the repertoire of accessible mRNA modifications; (2) synthesis strategies for linear, circular, and branched mRNA topologies; (3) structure–activity relationships governing translation efficiency, decay, and immune activation; and (4) implications for next-generation mRNA-based therapeutics. By integrating chemical synthesis, synthetic biology, and RNA structural design, researchers are beginning to unlock the full therapeutic potential of engineered mRNA molecules.

Current mechanical and chemical recycling strategies address less than 10% of global plastic waste, necessitating alternative valorization routes. Biological upcycling via enzymatic depolymerization combined with microbial conversion of the resulting monomers offers a promising pathway to transform mixed plastic waste into valuable alternatives. Here, we employed a single engineered Pseudomonas putida KT2440 for simultaneous co-utilization of five plastic monomers including ethylene glycol, terephthalic acid, adipic acid, 1,4-butanediol, and L-lactic acid, which can be derived from enzymatic hydrolysis of polyethylene terephthalate (PET), polybutylene adipate-co-terephthalate (PBAT), polyester-polyurethanes (PUs), and polylactic acid (PLA). Continuous fermentation over 21 days with alternating mixed-monomer feeds achieved steady state growth and complete substrate depletion, yielding adaptive mutations that informed iterative strain improvement. Further engineering enabled the biosynthesis of (R)-3-hydroxybutyrate (R-3HB), and 0.70 g/L R-3HB was produced directly from enzymatic hydrolysates of blended PET, PBAT, and TPU. These results establish a viable bio-based approach for upcycling realistic mixed plastics into value-added bioproducts.

Constructing and studying pangenome variation graphs (PVGs) supports new insights into viral genomic diversity. This is because such pangenomes are less prone to reference bias, which affects mutation detection. Interpreting the information arising from this is challenging, so automating these processes to allow exploratory investigations for PVG optimization is essential. Moreover, existing methods do not scale well to the smaller virus genome sizes and to facilitate analysis in laptop environments. To address this, we developed an easily deployable pipeline to facilitate the rapid creation of virus PVGs that applies a broad range of analyses to these PVGs.  We present Panalyze, a computationally scalable virus PVG construction, analysis and annotation tool implemented in NextFlow and containerised in Docker. Panalyze uses NextFlow to efficiently complete tasks across multiple compute nodes and in diverse computing environments. Panalyze can also operate on a single thread on a standard laptop, and analyse sequence lengths of any size. We illustrate how Panalyze works and the valuable outputs it can generate using a range of common viral pathogens.

Accurate prediction of protein function is essential for elucidating molecular mechanisms and advancing biological and therapeutic discovery. Yet experimental annotation lags far behind the rapid growth of protein sequence data. Computational approaches address this gap by associating proteins with Gene Ontology (GO) terms, which encode functional knowledge through hierarchical relations and textual definitions. However, existing models often emphasize one modality over the other, limiting their ability to generalize, particularly to unseen or newly introduced GO terms that frequently arise as the ontology evolves, and making the previously trained models outdated.  We present STAR-GO, a Transformer-based framework that jointly models the semantic and structural characteristics of GO terms to enhance zero-shot protein function prediction. STAR-GO integrates textual definitions with ontology graph structure to learn unified GO representations, which are processed in hierarchical order to propagate information from general to specific terms. These representations are then aligned with protein sequence embeddings to capture sequence–function relationships. STAR-GO achieves state-of-the-art performance and superior zero-shot generalization, demonstrating the utility of integrating semantics and structure for robust and adaptable protein function prediction.

Lig E is a periplasm-targeted ATP-dependent DNA ligase found in many Gram-negative bacteria including Neisseria gonorrhoeae. Although LigE has been shown to have a role in biofilm formation, many Lig E-possessing bacteria are also naturally competent, suggesting a possible function in transformation with extracellular DNA. Here, we demonstrate that Lig E participates in bacterial competence by increasing transformation with nuclease-damaged extracellular DNA that contains single-stranded or cohesive breaks. We show that increased transformation with this restricted DNA is ATP-dependent, and that the ATP concentration increases in the extracellular milieu during maintenance of N. gonorrhoeae in liquid culture.