New methods for antimicrobial design are critical for combating pathogenic bacteria in the post-antibiotic era. Fortunately, competition within complex communities has led to the natural evolution of antimicrobial peptide (AMP) sequences that have promising bactericidal properties. Unfortunately, the identification, characterization, and production of AMPs can prove complex and time consuming. Here, we report a peptide generation framework, PepVAE, based around variational autoencoder (VAE) and antimicrobial activity prediction models for designing novel AMPs using only sequences and experimental minimum inhibitory concentration (MIC) data as input. Sampling from distinct regions of the learned latent space allows for controllable generation of new AMP sequences with minimal input parameters. Extensive analysis of the PepVAE-generated sequences paired with antimicrobial activity prediction models supports this modular design framework as a promising system for development of novel AMPs, demonstrating controlled production of AMPs with experimental validation of predicted antimicrobial activity.

Electroanalytical techniques are useful for detection and identification because the instrumentation is simple and can support a wide variety of assays. One example is cyclic square wave voltammetry (CSWV), a practical detection technique for different classes of compounds including explosives, herbicides/pesticides, industrial compounds, and heavy metals. A key barrier to the widespread application of CSWV for chemical identification is the necessity of a high performance, generalizable classification algorithm. Here, machine and deep learning models were developed for classifying samples based on voltammograms alone. The highest performing models were Long Short-Term Memory (LSTM) and Fully Convolutional Networks (FCNs), depending on the dataset against which performance was assessed. When compared to other algorithms, previously used for classification of CSWV and other similar data, our LSTM and FCN-based neural networks achieve higher sensitivity and specificity with the area under the curve values from receiver operating characteristic (ROC) analyses greater than 0.99 for several datasets. Class activation maps were paired with CSWV scans to assist in understanding the decision-making process of the networks, and their ability to utilize this information was examined. The best-performing models were then successfully applied to new or holdout experimental data. An automated method for processing CSWV data, training machine learning models, and evaluating their prediction performance is described, and the tools generated provide support for the identification of compounds using CSWV from samples in the field.

Nanoparticle scaffolds can impart multiple benefits onto immobilized enzymes including enhanced stability, activity, and recoverability. The magnitude of these benefits is modulated by features inherent to the scaffold–enzyme conjugate, amongst which the size of the nanoscaffold itself can be critically important. In this review, we highlight the benefits of enzyme immobilization on nanoparticles and the factors affecting these benefits using quantum dots and gold nanoparticles as representative materials due to their maturity. We then review recent literature on the use of these scaffolds for enzyme immobilization and as a means to dissect the underlying mechanisms. Detailed analysis of the literature suggests that there is a “sweet-spot” for scaffold size and the ratio of immobilized enzyme to scaffold, with smaller scaffolds and lower enzyme:scaffold ratios generally providing higher enzymatic activities. We anticipate that ongoing studies of enzyme immobilization onto nanoscale scaffolds will continue to sharpen our understanding of what gives rise to beneficial characteristics and allow for the next important step, namely, that of translation to large-scale processes that exploit these properties.

Successful proteomic characterization of biological material depends on the development of robust sample processing methods. The acorn barnacle Amphibalanus amphitrite is a biofouling model for adhesive processes, but the identification of causative proteins involved has been hindered by their insoluble nature. Although effective, existing sample processing methods are labor and time intensive, slowing progress in this field. Here, a more efficient sample processing method is described which exploits pressure cycling technology (PCT) in combination with protein solvents. PCT aids in protein extraction and digestion for proteomics analysis. Barnacle adhesive proteins can be extracted and digested in the same tube using PCT, minimizing sample loss, increasing throughput to 16 concurrently processed samples, and decreasing sample processing time to under 8 hours. PCT methods produced similar proteomes in comparison to previous methods. Two solvents which were ineffective at extracting proteins from the adhesive at ambient pressure (urea and methanol) produced more protein identifications under pressure than highly polar hexafluoroisopropanol, leading to the identification and description of >40 novel proteins at the interface. Some of these have homology to proteins with elastomeric properties or domains involved with protein-protein interactions, while many have no sequence similarity to proteins in publicly available databases, highlighting the unique adherent processes evolved by barnacles. The methods described here can not only be used to further characterize barnacle adhesive to combat fouling, but may also be applied to other recalcitrant biological samples, including aggregative or fibrillar protein matrices produced during disease, where a lack of efficient sample processing methods has impeded advancement. Data are available via ProteomeXchange with identifier PXD012730.

Single-domain antibodies (sdAbs) or nanobodies have received widespread attention due to their small size (~ 15 kDa) and diverse applications in bio-derived therapeutics. As many modern biotechnology breakthroughs are applied to antibody engineering and design, nanobody thermostability or melting temperature (T_m) is crucial for their successful utilization. In this study, we present TEMPRO which is a predictive modeling approach for estimating the T_m of nanobodies using computational methods. Our methodology integrates various nanobody biophysical features to include Evolutionary Scale Modeling (ESM) embeddings, NetSurfP3 structural predictions, pLDDT scores per sdAb region from AlphaFold2, and each sequence’s physicochemical characteristics. This approach is validated with our combined dataset containing 567 unique sequences with corresponding experimental T_m values from a manually curated internal data and a recently published nanobody database, NbThermo. Our results indicate the efficacy of protein embeddings in reliably predicting the T_m of sdAbs with mean absolute error (MAE) of 4.03 °C and root mean squared error (RMSE) of 5.66 °C, thus offering a valuable tool for the optimization of nanobodies for various biomedical and therapeutic applications. Moreover, we have validated the models’ performance using experimentally determined T_ms from nanobodies not found in NbThermo. This predictive model not only enhances nanobody thermostability prediction, but also provides a useful perspective of using embeddings as a tool for facilitating a broader applicability of downstream protein analyses.

The electrochemical detection of heavy metal ions is reported using an inexpensive portable in-house built potentiostat and epitaxial graphene. Monolayer, hydrogen-intercalated quasi-freestanding bilayer, and multilayer epitaxial graphene were each tested as working electrodes before and after modification with an oxygen plasma etch to introduce oxygen chemical groups to the surface. The graphene samples were characterized using X-ray photoelectron spectroscopy, atomic force microscopy, Raman spectroscopy, and van der Pauw Hall measurements. Dose–response curves in seawater were evaluated with added trace levels of four heavy metal salts (CdCl₂, CuSO₄, HgCl₂, and PbCl₂), along with detection algorithms based on machine learning and library development for each form of graphene and its oxygen plasma modification. Oxygen plasma-modified, hydrogen-intercalated quasi-freestanding bilayer epitaxial graphene was found to perform best for correctly identifying heavy metals in seawater.