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 (Tm) is crucial for their successful utilization. In this study, we present TEMPRO which is a predictive modeling approach for estimating the Tm 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 Tm 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 Tm 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 Tms 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.

Efficient and accurate classification of DNA barcode data is crucial for large-scale fungal biodiversity studies. However, existing methods are either computationally expensive or lack accuracy. Previous research has demonstrated the potential of deep learning in this domain, successfully training neural networks for biological sequence classification. We introduce the MycoAI Python package, featuring various deep learning models such as BERT and CNN tailored for fungal Internal Transcribed Spacer (ITS) sequences. We explore different neural architecture designs and encoding methods to identify optimal models. By employing a multi-head output architecture and multi-level hierarchical label smoothing, MycoAI effectively generalizes across the taxonomic hierarchy. Using over 5 million labelled sequences from the UNITE database, we develop two models: MycoAI-BERT and MycoAI-CNN. While we emphasize the necessity of verifying classification results by AI models due to insufficient reference data, MycoAI still exhibits substantial potential. When benchmarked against existing classifiers such as DNABarcoder and RDP on two independent test sets with labels present in the training dataset, MycoAI models demonstrate high accuracy at the genus and higher taxonomic levels, with MycoAI-CNN being the fastest and most accurate. In terms of efficiency, MycoAI models can classify over 300,000 sequences within 5 min. We publicly release the MycoAI models, enabling mycologists to classify their ITS barcode data efficiently. Additionally, MycoAI serves as a platform for developing further deep learning-based classification methods. The source code for MycoAI is available under the MIT Licence at https://github.com/MycoAI/MycoAI.

Owing to the overlapping and cross-interference of absorption lines in multicomponent gases, the simultaneous measurement of such gases via laser absorption spectroscopy frequently necessitates the use of supplementary pressure sensors to distinguish the spectral lines. Alternatively, it requires multiple lasers combined with time-division multiplexing to independently scan the absorption peaks of each gas, thereby preventing interference from other gases. This inevitably escalates both the cost of the system and the complexity of the gas pathway. In response to these challenges, a mid-infrared sensor employing a neural network-based decoupling algorithm for aliasing spectral is developed, enabling the simultaneous detection of methane(CH4), water vapor(H2O), and ethane(C2H6). The sensor system underwent evaluation in a controlled laboratory environment. Allan deviation analysis revealed that the minimum detection limits for CH4,H2O, and C2H6 were 6.04, 118.44, and 1 ppb, respectively, with an averaging time of 3 s. The performance of the proposed sensor demonstrates that the aliasing spectral decoupling algorithm based on neural network combined with wavelength-modulated spectroscopy technology has the advantages of high sensitivity, low cost and low complexity, showing its potential for simultaneous detection of multicomponent trace gases in various fields.

Stabilized enzymes are crucial for the industrial application of biocatalysis due to their enhanced operational stability, which leads to prolonged enzyme activity, cost-efficiency and consequently scalability of biocatalytic processes. Over the past decade, numerous studies have demonstrated that deep eutectic solvents (DES) are excellent enzyme stabilizers. However, the search for an optimal DES has primarily relied on trial-and-error methods, lacking systematic exploration of DES structure-activity relationships. Therefore, this study aims to rationally design DES to stabilize various dehydrogenases through extensive experimental screening, followed by the development of a straightforward and reliable mathematical model to predict the efficacy of DES in enzyme stabilization. A total of 28 DES were tested for their ability to stabilize three dehydrogenases at 30°C: (S)-alcohol dehydrogenase from Rhodococcus ruber (ADH-A), (R)-alcohol dehydrogenase from Lactobacillus kefir (Lk-ADH) and glucose dehydrogenase from Bacillus megaterium (GDH). The residual activity of these enzymes in the presence of DES was quantified using first-order kinetic models. The screening revealed that DES based on polyols serve as promising stabilizing environments for the three tested dehydrogenases, particularly for the enzymes Lk-ADH and GDH, which are intrinsically unstable in aqueous environments. In glycerol-based DES, increases in enzyme half-life of up to 175-fold for Lk-ADH and 60-fold for GDH were observed compared to reference buffers. Furthermore, to establish the relationship between the enzyme inactivation rate constants and DES descriptors generated by the Conductor-like Screening Model for Real Solvents, artificial neural network models were developed. The models for ADH-A and GDH showed high efficiency and reliability (R2 > 0.75) for in silico screening of the enzyme inactivation rate constants based on DES descriptors. In conclusion, these results highlight the significant potential of the integrated experimental and in silico approach for the rational design of DES tailored to stabilize enzymes.

Globalization has led to the frequent movement of species out of their native habitat. Some of these species become highly invasive and capable of profoundly altering invaded ecosystems. Feral swine (Sus scrofa × domesticus) are recognized as being among the most destructive invasive species, with populations established on all continents except Antarctica. Within the United States (US), feral swine are responsible for extensive crop damage, the destruction of native ecosystems, and the spread of disease. Purposeful human-mediated movement of feral swine has contributed to their rapid range expansion over the past 30 years. Patterns of deliberate introduction of feral swine have not been well described as populations may be established or augmented through small, undocumented releases. By leveraging an extensive genomic database of 18,789 samples genotyped at 35,141 single nucleotide polymorphisms (SNPs), we used deep neural networks to identify translocated feral swine across the contiguous US. We classified 20% (3364/16,774) of sampled animals as having been translocated and described general patterns of translocation using measures of centrality in a network analysis. These findings unveil extensive movement of feral swine well beyond their dispersal capabilities, including individuals with predicted origins >1000 km away from their sampling locations. Our study provides insight into the patterns of human-mediated movement of feral swine across the US and from Canada to the northern areas of the US. Further, our study validates the use of neural networks for studying the spread of invasive species.

UNASSIGNED: Cancer remains one of the leading causes of mortality globally, with conventional chemotherapy often resulting in severe side effects and limited effectiveness. Recent advancements in bioinformatics and machine learning, particularly deep learning, offer promising new avenues for cancer treatment through the prediction and identification of anticancer peptides.
UNASSIGNED: This study aimed to develop and evaluate a deep learning model utilizing a two-dimensional convolutional neural network (2D CNN) to enhance the prediction accuracy of anticancer peptides, addressing the complexities and limitations of current prediction methods.
UNASSIGNED: A diverse dataset of peptide sequences with annotated anticancer activity labels was compiled from various public databases and experimental studies. The sequences were preprocessed and encoded using one-hot encoding and additional physicochemical properties. The 2D CNN model was trained and optimized using this dataset, with performance evaluated through metrics such as accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC-ROC).
UNASSIGNED: The proposed 2D CNN model achieved superior performance compared to existing methods, with an accuracy of 0.87, precision of 0.85, recall of 0.89, F1-score of 0.87, and an AUC-ROC value of 0.91. These results indicate the model\'s effectiveness in accurately predicting anticancer peptides and capturing intricate spatial patterns within peptide sequences.
UNASSIGNED: The findings demonstrate the potential of deep learning, specifically 2D CNNs, in advancing the prediction of anticancer peptides. The proposed model significantly improves prediction accuracy, offering a valuable tool for identifying effective peptide candidates for cancer treatment.
UNASSIGNED: Further research should focus on expanding the dataset, exploring alternative deep learning architectures, and validating the model\'s predictions through experimental studies. Efforts should also aim at optimizing computational efficiency and translating these predictions into clinical applications.

UNASSIGNED: Recent research explores using neural networks to reconstruct undersampled magnetic resonance imaging. Because of the complexity of the artifacts in the reconstructed images, there is a need to develop task-based approaches to image quality. We compared conventional global quantitative metrics to evaluate image quality in undersampled images generated by a neural network with human observer performance in a detection task. The purpose is to study which acceleration (2×, 3×, 4×, 5×) would be chosen with the conventional metrics and compare it to the acceleration chosen by human observer performance.
UNASSIGNED: We used common global metrics for evaluating image quality: the normalized root mean squared error (NRMSE) and structural similarity (SSIM). These metrics are compared with a measure of image quality that incorporates a subtle signal for a specific task to allow for image quality assessment that locally evaluates the effect of undersampling on a signal. We used a U-Net to reconstruct under-sampled images with 2×, 3×, 4×, and 5× one-dimensional undersampling rates. Cross-validation was performed for a 500- and a 4000-image training set with both SSIM and MSE losses. A two-alternative forced choice (2-AFC) observer study was carried out for detecting a subtle signal (small blurred disk) from images with the 4000-image training set.
UNASSIGNED: We found that for both loss functions, the human observer performance on the 2-AFC studies led to a choice of a 2× undersampling, but the SSIM and NRMSE led to a choice of a 3× undersampling.
UNASSIGNED: For this detection task using a subtle small signal at the edge of detectability, SSIM and NRMSE led to an overestimate of the achievable undersampling using a U-Net before a steep loss of image quality between 2×, 3×, 4×, 5× undersampling rates when compared to the performance of human observers in the detection task.

A main goal in neuroscience is to understand the computations carried out by neural populations that give animals their cognitive skills. Neural network models allow to formulate explicit hypotheses regarding the algorithms instantiated in the dynamics of a neural population, its firing statistics, and the underlying connectivity. Neural networks can be defined by a small set of parameters, carefully chosen to procure specific capabilities, or by a large set of free parameters, fitted with optimization algorithms that minimize a given loss function. In this work we alternatively propose a method to make a detailed adjustment of the network dynamics and firing statistic to better answer questions that link dynamics, structure, and function. Our algorithm-termed generalised Firing-to-Parameter (gFTP)-provides a way to construct binary recurrent neural networks whose dynamics strictly follows a user pre-specified transition graph that details the transitions between population firing states triggered by stimulus presentations. Our main contribution is a procedure that detects when a transition graph is not realisable in terms of a neural network, and makes the necessary modifications in order to obtain a new transition graph that is realisable and preserves all the information encoded in the transitions of the original graph. With a realisable transition graph, gFTP assigns values to the network firing states associated with each node in the graph, and finds the synaptic weight matrices by solving a set of linear separation problems. We test gFTP performance by constructing networks with random dynamics, continuous attractor-like dynamics that encode position in 2-dimensional space, and discrete attractor dynamics. We then show how gFTP can be employed as a tool to explore the link between structure, function, and the algorithms instantiated in the network dynamics.

This review aims to assess the effectiveness of AI-driven CDSSs on patient outcomes and clinical practices. A comprehensive search was conducted across PubMed, MEDLINE, and Scopus. Studies published from January 2018 to November 2023 were eligible for inclusion. Following title and abstract screening, full-text articles were assessed for methodological quality and adherence to inclusion criteria. Data extraction focused on study design, AI technologies employed, reported outcomes, and evidence of AI-CDSS impact on patient and clinical outcomes. Thematic analysis was conducted to synthesise findings and identify key themes regarding the effectiveness of AI-CDSS. The screening of the articles resulted in the selection of 26 articles that satisfied the inclusion criteria. The content analysis revealed four themes: early detection and disease diagnosis, enhanced decision-making, medication errors, and clinicians\' perspectives. AI-based CDSSs were found to improve clinical decision-making by providing patient-specific information and evidence-based recommendations. Using AI in CDSSs can potentially improve patient outcomes by enhancing diagnostic accuracy, optimising treatment selection, and reducing medical errors.

The access to increasing volumes of scientific and clinical data, particularly with the implementation of electronic health records, has reignited an enthusiasm for artificial intelligence and its application to the health sciences. This interest has reached a crescendo in the past few years with the development of several machine learning- and deep learning-based medical technologies. The impact on research and clinical practice within gastroenterology and hepatology has already been significant, but the near future promises only further integration of artificial intelligence and machine learning into this field. The concepts underlying artificial intelligence and machine learning initially seem intimidating, but with increasing familiarity, they will become essential skills in every clinician\'s toolkit. In this review, we provide a guide to the fundamentals of machine learning, a concentrated area of study within artificial intelligence that has been built on a foundation of classical statistics. The most common machine learning methodologies, including those involving deep learning, are also described.