J Affect Disord. 2025 Mar 11:S0165-0327(25)00370-2. doi: 10.1016/j.jad.2025.03.032. Online ahead of print.

ABSTRACT

BACKGROUND: Recent studies have linked branched-chain amino acids (BCAAs) metabolism with the risk of major depressive disorder (MDD). However, it is unclear whether associations of plasma BCAA levels with MDD are causal or driven by reverse causality.

METHODS: Mendelian randomization (MR) was used to investigate the causal association of genetically determined BCAA levels with the risk of MDD. The large genome-wide association study (GWAS) datasets on plasma BCAA levels (n = 115,051) were obtained from the UK Biobank. The summary GWAS dataset for MDD was obtained from the Psychiatric Genomics Consortium (n = 1,035,760). We applied the inverse variance-weighted (IVW) method to explore the causal relationships between BCAA levels and MDD, followed by multiple pleiotropy and heterogeneity tests.

RESULTS: Our results demonstrated that genetically determined circulating total BCAAs (odds ratio (OR): 1.05, 95 % confidence interval (CI): 1.01-1.10, P = 0.016), leucine (OR: 1.06, 95 % CI: 1.02-1.11, P = 7.22 × 10-3), and isoleucine (OR: 1.08, 95 % CI: 1.01-1.16, P = 0.032) levels were associated with an increased risk of MDD. There was suggestive evidence supporting the causal effect of valine levels on MDD (OR: 1.04, 95 % CI: 1.00-1.08, P = 0.075). Bidirectional MR analysis did not provide evidence of reverse causality.

CONCLUSIONS: We report evidence supporting the causal role of BCAAs in the development of MDD. This study offers new insights into the mechanisms and treatment of MDD.

PMID:40081595 | DOI:10.1016/j.jad.2025.03.032

Metab Eng. 2025 Mar 11:S1096-7176(25)00033-3. doi: 10.1016/j.ymben.2025.03.003. Online ahead of print.

ABSTRACT

Microbial poly(3-hydroxybutyrate-co-4-hydroxybutyrate-co-3-hydroxyvalerate), abbreviated as P(3HB-4HB-3HV) or P34HBHV, is a flexible polyhydroxyalkanoate (PHA) material ranging from softness to elasticity depending on the ratios of various monomers. Halomonas bluephagenesis, as the chassis of the next generation industrial biotechnology (NGIB) able to grow contamination free under open unsterile conditions. The resulting recombinants of H. bluephagenesis became capable of efficiently synthesizing P34HBHV utilizing glucose as the sole carbon source. Engineered H. bluephagenesis H1 (encoding ogdA, sucD, 4hbD, orfZ, scpA and scpB in chromosomes) transformed with a plasmid containing PHA synthesis genes phaC and phaA and its derivative H29 produced up to 92% P(3HB-co-8.85%4HB-co-8.47%3HV) and 72% P(3HB-co-13.21%4HB-co-11.97%3HV) in cell dry weight (CDW), respectively, in shake flasks. In bioreactor cultivation, H. bluephagenesis H39 constructed by integrating the 4hbD, phaC and phaA genes into the genome of H. bluephagenesis H1 achieved 95 g/L CDW with 69% P(3HB-co-10.49%4HB-co-3.54%3HV), while H. bluephagenesis H43, further optimized with lpxM deletion, reached 73 g/L CDW with 78% P(3HB-co-10.35%4HB-co-4.54%3HV) in a 100 L bioreactor. For the first time, H. bluephagenesis was successfully engineered to generate stable and hyperproductive derivative strains for pilot production of P(3HB-4HB-3HV) with customizable monomer ratios from glucose as the sole carbon source.

PMID:40081465 | DOI:10.1016/j.ymben.2025.03.003

Metab Eng. 2025 Mar 11:S1096-7176(25)00028-X. doi: 10.1016/j.ymben.2025.02.015. Online ahead of print.

ABSTRACT

Advances in genomics technologies have generated large data sets that provide tremendous insights into the genetic diversity of taxonomic groups. However, it remains challenging to pinpoint the effect of genetic diversity on different traits without performing resource-intensive phenotyping experiments. Pan-genome-scale metabolic models (panGEMs) extend traditional genome-scale metabolic models by considering the entire reaction repertoire that enables the prediction and comparison of metabolic capabilities within a taxonomic group. Here, we systematically review the state-of-the-art methodologies for constructing panGEMs, focusing on used tools, databases, experimental datasets, and orthology relationships. We highlight the unique advantages of panGEMs compared to single-species GEMs in predicting metabolic phenotypes and in guiding the experimental validation of genome annotations. In addition, we emphasize the disparity between the available (pan-)genomic data on photosynthetic organisms and their under-representation in current (pan)GEMs. Finally, we propose a perspective for tackling the reconstruction of panGEMs for photosynthetic eukaryotes that can help advance our understanding of the metabolic diversity in this taxonomic group.

PMID:40081464 | DOI:10.1016/j.ymben.2025.02.015

Semin Liver Dis. 2025 Mar 13. doi: 10.1055/a-2516-2361. Online ahead of print.

ABSTRACT

Acute-on-chronic liver failure (ACLF) is a severe condition in patients with decompensated liver cirrhosis, marked by high short-term mortality. Recent experimental and clinical evidence has linked intestinal dysfunction to both the initiation of ACLF as well as disease outcome. This review discusses the significant role of the gut-liver axis in ACLF pathogenesis, highlighting recent advances. Gut mucosal barrier disruption, gut dysbiosis, and bacterial translocation emerge as key factors contributing to systemic inflammation in ACLF. Different approaches of therapeutically targeting the gut-liver axis via farnesoid X receptor agonists, nonselective beta receptor blockers, antibiotics, and probiotics are discussed as potential strategies mitigating ACLF progression. The importance of understanding the distinct pathophysiology of ACLF compared with other stages of liver cirrhosis is highlighted. In conclusion, research findings suggest that disruption of intestinal integrity may be an integral component of ACLF pathogenesis, paving the way for novel diagnostic and therapeutic approaches to manage this syndrome more effectively.

PMID:40081417 | DOI:10.1055/a-2516-2361

J Pharmacol Exp Ther. 2025 Feb 21;392(4):103403. doi: 10.1016/j.jpet.2025.103403. Online ahead of print.

ABSTRACT

Attention deficit hyperactivity disorder (ADHD) is a neurodevelopmental disorder with unclear pathological mechanisms. ADHD is treated with both stimulant and nonstimulant medications, but their therapeutic mechanisms and impact on brain metabolites are not fully understood. This study employed an untargeted metabolomics approach with liquid chromatography mass spectrometry to investigate the pathogenesis of ADHD, as well as the effects of established and novel therapeutics. We characterized the metabolomic signatures of the adgrl3.1 mutant zebrafish ADHD model and examined the impact of methylphenidate, guanfacine, atomoxetine, and 5 novel putative therapeutics identified in a prior screen, including amlodipine. Our analysis revealed that the drugs commonly affect pathways related to amino acid and lipid metabolism, specifically involving glycine, serine, threonine, phenylalanine, lysophosphatidylcholine, and sphingomyelin. This convergence on similar metabolic targets was unexpected and suggests a broader, systemic effect of ADHD therapeutics, challenging the traditional view of distinct drug mechanisms. Amlodipine exhibited metabolic effects consistent with established treatments, indicating its potential as a viable alternative or adjunct therapy. These findings provide new insights into the metabolic underpinnings of ADHD and highlight potential targets for developing improved therapeutic strategies. SIGNIFICANCE STATEMENT: This study explores the metabolic pathways affected by attention deficit hyperactivity disorder treatments using a zebrafish adgrl3.1 mutant model. Untargeted metabolomics revealed that both established and novel attention deficit hyperactivity disorder medications influence common amino acid and lipid metabolism pathways, suggesting systemic effects. Notably, amlodipine showed similar impacts as current drugs, offering promise as an alternative therapy.

PMID:40081232 | DOI:10.1016/j.jpet.2025.103403

J Phys Chem B. 2025 Mar 13. doi: 10.1021/acs.jpcb.4c08640. Online ahead of print.

ABSTRACT

The dynamic formation of (bio)molecular condensates has emerged as a key regulatory mechanism in cellular processes. Concepts from polymer physics can provide valuable insights into the underlying mechanisms and properties of these condensates. While stoichiometric interactions between chemically distinct molecules have traditionally been the primary focus for understanding and predicting the equilibrium behavior, recent attention has turned to the role of molecular diversity, particularly the interplay between molecules of similar types but varying chain lengths. To mimic such cellular conditions, we investigated the impact of molecular weight polydispersity using polyethylene glycol (PEG) and dextran (Dex) solutions through experiments and molecular simulations. Our findings reveal that polydisperse systems, which contain a higher fraction of short-chain components, exhibit a narrower two-phase region, along with reduced concentration differences and interfacial tension between the coexisting polymer-rich and polymer-poor phases. In these systems, the Dex-rich phase is enriched with longer Dex chains compared to the PEG-rich phase, with a gradual transition in chain length across their interface. However, polydispersity has no significant effects on the critical concentration and critical exponents. Finally, our study of condensation kinetics demonstrates that phase separation is not limited by the nucleation rate but instead by the diffusion-driven aggregation of polymers.

PMID:40080692 | DOI:10.1021/acs.jpcb.4c08640

ACS Synth Biol. 2025 Mar 13. doi: 10.1021/acssynbio.4c00806. Online ahead of print.

ABSTRACT

Interest in Xanthobacter species is increasing due to their unique metabolic capabilities. They can grow in both heterotrophic and fully autotrophic environments, including carbon dioxide, dinitrogen gas, and hydrogen as the sole carbon, nitrogen, and energy sources, respectively. Academic and industrial groups looking to leverage these metabolic properties are already using Xanthobacter strains for the sustainable production of food and commodities. However, only a handful of genetic parts and protocols exist in scattered genetic backgrounds, and there is an unmet need for reliable genetic engineering tools to manipulate Xanthobacter species. Here, we developed XanthoMoClo, a robust modular cloning genetic toolkit for Xanthobacter and Roseixanthobacter species and strains, providing extensive tools to transform them, manipulate their metabolism, and express genes of interest. The toolkit contains plasmid parts, such as replication origins, antibiotic selection markers, fluorescent proteins, constitutive and inducible promoters, a standardized framework to incorporate novel components into the toolkit, and a conjugation donor to transform Xanthobacter and Roseixanthobacter strains easily with no or minimal optimization. We validated these plasmid components in depth in three of the most commonly studied Xanthobacter strains: X. versatilis Py2, X. autotrophicus GZ29, and X. flavus GJ10, as well as in R. finlandensis VTT E-85241. Finally, we demonstrate robust toolkit functionality across 21 different species of Xanthobacter and Roseixanthobacter, comprising 23 strains in total. The XanthoMoClo genetic toolkit is available to the research community (through AddGene) and will help accelerate the genetic engineering of Xanthobacter to further their applications in sustainability and bioremediation efforts.

PMID:40080684 | DOI:10.1021/acssynbio.4c00806

Hum Nat. 2025 Mar 13. doi: 10.1007/s12110-025-09488-8. Online ahead of print.

ABSTRACT

Automatic behavioral matching includes Rapid Facial Mimicry (RFM) and Yawn Contagion (YC) that occur when the facial expression of an individual acts as a 'mirror social releaser' and induces the same facial expression in the observer (within 1 s for RFM, and minutes for YC). Motor replication has been linked to coordination and emotional contagion, a basic form of empathy. We investigated the presence and modulating factors of Rapid Smile Mimicry (RSM) and YC in infants/toddlers from 10 to 36 months at the nursery 'Melis' (Turin, Italy). In February-May 2022, we gathered audio and/or video of all occurrences data on affiliative behaviors, smiling during play, and yawning during everyday activities. Both RSM and YC were present, as toddlers were most likely to smile (within 1 s) or yawn (within three-min) after perceiving a smile/yawn from another toddler. Sex, age, and parents' country of origin did not influence RSM and YC occurrence, probably because gonadal maturation was long to come, the age range was skewed towards the early developmental phase, and toddlers had been in the same social group for months. RSM and YC showed social modulation, thus possibly implying more than just motor resonance. Both phenomena were inversely related to affiliation levels (a social bond proxy). Because literature reports that in adults RSM and YC may increase with familiarity, our reversed result suggests that in certain toddler cohorts the same phenomena may help increase socio-emotional coordination and that the function of motoric resonance may be experience- and context-dependent.

PMID:40080328 | DOI:10.1007/s12110-025-09488-8

J Mol Neurosci. 2025 Mar 13;75(1):32. doi: 10.1007/s12031-025-02328-5.

ABSTRACT

Caloric restriction (CR) is a dietary intervention that reduces calorie intake without inducing malnutrition, demonstrating lifespan-extending effects in preclinical studies and some human trials, along with potential benefits in ameliorating age-related ailments. Caloric restriction mimetics (CRMs) are compounds mimicking CR effects, offering a potential therapeutic avenue for age-related diseases. This study explores the potential protective effects of CR on the brain neocortex (GSE11291) and the identification of CRMs using integrative bioinformatics and systems biology approaches. Our findings indicate that long-term CR activates cellular pathways improving mitochondrial function, enhancing antioxidant capacity, and reducing inflammation, potentially providing neuroprotection. The key signaling pathways enriched in our study include PPAR, mTOR, FoxO, AMPK, and Notch signaling pathways, which are crucial regulators of metabolism, cellular stress response, neuroprotection, and longevity. We identify key signaling molecules and molecular mechanisms associated with CR, including transcription factors, kinase regulators, and microRNAs linked to differentially expressed genes. Furthermore, potential CRMs such as rapamycin, replicating CR-related health benefits, are identified. Additionally, machine learning models were developed to classify small molecules based on their CNS activity and anti-inflammatory properties. As a proof of concept, we have demonstrated the ischemic neuroprotective effects of two top-ranked candidate reference molecules (CRMs) using the oxygen-glucose deprivation (OGD) model, an established in vitro stroke model. However, further investigations are essential to fully elucidate the therapeutic potential of these CRMs. In summary, our study suggests that long-term CR entails protective mechanisms preserving and safeguarding neuronal function, potentially impacting the treatment of age-related neurological diseases. Moreover, our findings contribute to the identification of potential genes and regulatory molecules involved in CR, along with potential CRMs, providing a promising foundation for future research in the field of neurological disorder treatment.

PMID:40080242 | DOI:10.1007/s12031-025-02328-5

J Proteome Res. 2025 Mar 13. doi: 10.1021/acs.jproteome.4c00864. Online ahead of print.

ABSTRACT

Mass spectrometry-based discovery of bacterial immunopeptides presented by infected cells allows untargeted discovery of bacterial antigens that can serve as vaccine candidates. However, reliable identification of bacterial epitopes is challenged by their extremely low abundance. Here, we describe an optimized bioinformatic framework to enhance the confident identification of bacterial immunopeptides. Immunopeptidomics data of cell cultures infected with Listeria monocytogenes were searched by four different search engines, PEAKS, Comet, Sage and MSFragger, followed by data-driven rescoring with MS2Rescore. Compared with individual search engine results, this integrated workflow boosted immunopeptide identification by an average of 27% and led to the high-confidence detection of 18 additional bacterial peptides (+27%) matching 15 different Listeria proteins (+36%). Despite the strong agreement between the search engines, a small number of spectra (<1%) had ambiguous matches to multiple peptides and were excluded to ensure high-confidence identifications. Finally, we demonstrate our workflow with sensitive timsTOF SCP data acquisition and find that rescoring, now with inclusion of ion mobility features, identifies 76% more peptides compared to Q Exactive HF acquisition. Together, our results demonstrate how integration of multiple search engine results along with data-driven rescoring maximizes immunopeptide identification, boosting the detection of high-confidence bacterial epitopes for vaccine development.

PMID:40080147 | DOI:10.1021/acs.jproteome.4c00864

Drug Ther Bull. 2025 Mar 13:dtb-2025-000010. doi: 10.1136/dtb.2025.000010. Online ahead of print.

NO ABSTRACT

PMID:40081948 | DOI:10.1136/dtb.2025.000010

J Exp Med. 2025 Jun 2;222(6):e20230803. doi: 10.1084/jem.20230803. Epub 2025 Mar 13.

ABSTRACT

To distinguish pathogens from commensals, the intestinal epithelium employs cytosolic innate immune sensors. Activation of the NAIP-NLRC4 inflammasome initiates extrusion of infected intestinal epithelial cells (IEC) upon cytosolic bacterial sensing. We previously reported that activation of the inflammasome in tuft cells, which are primarily known for their role in parasitic infections, leads to the release of prostaglandin D2 (PGD2). We observe that NAIP-NLRC4 inflammasome activation in tuft cells leads to an antibacterial response with increased IL-22 and antimicrobial protein levels within the small intestine, which is dependent on PGD2 signaling. A NKp46+ subset of ILC3 expresses the PGD2 receptor CRTH2 and is the source of the increased IL-22. Inflammasome activation in tuft cells also leads to better control of Salmonella Typhimurium in the distal small intestine. However, tuft cells in the cecum and colon are dispensable for antibacterial immunity. These data support that intestinal tuft cells can also induce antibacterial responses, possibly in a tissue-specific manner.

PMID:40079814 | DOI:10.1084/jem.20230803

J Antimicrob Chemother. 2025 Mar 13:dkaf073. doi: 10.1093/jac/dkaf073. Online ahead of print.

ABSTRACT

BACKGROUND: Mycobacterium abscessus is an important cause of pulmonary infections, particularly among people with cystic fibrosis. Current treatment options for M. abscessus are suboptimal. Apramycin is a promising alternative aminoglycoside for M. abscessus, in part due to its ability to avoid intrinsic aminoglycoside-modifying enzymes in this pathogen.

OBJECTIVES: Define the pharmacodynamic activity of apramycin doses against M. abscessus.

METHODS: Apramycin and amikacin pharmacodynamics were assessed against two amikacin-susceptible M. abscessus subsp. abscessus isolates (ATCC 19977 and NR-44261) using a 14-day hollow fibre infection model (HFIM). Viable bacterial counts were determined during exposure to amikacin (15-20 mg/kg q24h) and 3 fractionated doses of apramycin (15 mg/kg q12h, 30 mg/kg q24h, 60 mg/kg q48h) using pharmacokinetic profiles predicted in epithelial lining fluid.

RESULTS: Against ATCC 19977, apramycin activity exceeded that of amikacin, with maximum bacterial reductions between 1.51 and 2.18 log10 cfu/mL for the different doses. Apramycin 15 mg/kg q12h displayed slightly better killing compared with the other apramycin dosing regimens between 96 and 144h before regrowth occurred. NR-44261 was not inhibited by amikacin and the activity of apramycin against this isolate was similar between the three doses (∼0.5 log10 cfu/mL reductions). After 14 days of exposure to apramycin monotherapy, ATCC 19977 and NR-44261 became apramycin resistant with MICs of >32 mg/L.

CONCLUSIONS: Apramycin exhibited greater pharmacodynamic activity than amikacin against amikacin-susceptible M. abscessus isolates and may be a promising therapy for this pathogen. However, antibiotic combination strategies to minimize apramycin resistance from emerging may be necessary.

PMID:40079270 | DOI:10.1093/jac/dkaf073

Acta Oncol. 2025 Mar 13;64:391-405. doi: 10.2340/1651-226X.2025.42982.

ABSTRACT

BACKGROUND AND PURPOSE: This study aims to develop and compare combined models based on enhanced CT-based radiomics, multi-dimensional deep learning, clinical-conventional imaging and spatial habitat analysis to achieve accurate prediction of thymoma risk classification.

MATERIALS AND METHODS: 205 consecutive patients with thymoma confirmed by surgical pathology were recruited from three medical centers. Venous phase enhanced CT images were used to delineate the tumor, and radiomics, 2D and 3D deep learning models based on the whole tumor were established and feature extraction was performed. The tumors were divided into different sub-regions by K-means clustering method and the corresponding features were obtained. The clinical-conventional imaging data of the patients were collected and evaluated, and the univariate and multivariate analysis were used for screening. The above types of features were fused with each other to construct a variety of combined models. Quantitative indicators such as area under the receiver operating characteristic (ROC) curve (AUC) were calculated to evaluate the performance of the model.

RESULTS: The AUC of RDLCSM developed based on LightGBM classifier was 0.953 in the training cohort, 0.930 in the internal validation cohort, 0.924 and 0.903 in the two external validation cohorts, respectively. RDLCSM performs better than RDLM (AUC range: 0.831-0.890) and 2DLCSM (AUC range: 0.785-0.916) based on KNN. In addition, RDLCSM had the highest accuracy (0.818-0.882) and specificity (0.926-1.000).

INTERPRETATION: The RDLCSM, which combines whole-tumor radiomics, 2D and 3D deep learning, clinical-visual radiology, and subregional omics, can be used as a non-invasive tool to predict thymoma risk classification.

PMID:40079653 | DOI:10.2340/1651-226X.2025.42982

Elife. 2025 Mar 13;13:RP97330. doi: 10.7554/eLife.97330.

ABSTRACT

Antimicrobial peptides (AMPs) are attractive candidates to combat antibiotic resistance for their capability to target biomembranes and restrict a wide range of pathogens. It is a daunting challenge to discover novel AMPs due to their sparse distributions in a vast peptide universe, especially for peptides that demonstrate potencies for both bacterial membranes and viral envelopes. Here, we establish a de novo AMP design framework by bridging a deep generative module and a graph-encoding activity regressor. The generative module learns hidden 'grammars' of AMP features and produces candidates sequentially pass antimicrobial predictor and antiviral classifiers. We discovered 16 bifunctional AMPs and experimentally validated their abilities to inhibit a spectrum of pathogens in vitro and in animal models. Notably, P076 is a highly potent bactericide with the minimal inhibitory concentration of 0.21 μM against multidrug-resistant Acinetobacter baumannii, while P002 broadly inhibits five enveloped viruses. Our study provides feasible means to uncover the sequences that simultaneously encode antimicrobial and antiviral activities, thus bolstering the function spectra of AMPs to combat a wide range of drug-resistant infections.

PMID:40079572 | DOI:10.7554/eLife.97330

Proc Inst Mech Eng H. 2025 Mar 13:9544119251325331. doi: 10.1177/09544119251325331. Online ahead of print.

ABSTRACT

The aim of this study is to develop Deep Learning (DL) enabled robotic systems to identify audio-based emotional pragmatics deficits in individuals with social pragmatic communication deficits. The novelty of the work stems from its integration of deep learning with a robotics platform for identifying emotional pragmatics deficits. In this study, the proposed methodology utilizes the implementation of machine and DL-based classification techniques, which have been applied to a collection of open-source datasets to identify audio emotions. The application of pre-processing and converting audio signals of different emotions utilizing Mel-Frequency Cepstral Coefficients (MFCC) resulted in improved emotion classification. The data generated using MFCC were used for the training of machine or DL models. The trained models were then tested on a randomly selected dataset. DL has been proven to be more effective in the identification of emotions using robotic structure. As the data generated by MFCC is of a single dimension, therefore, one-dimensional DL algorithms, such as 1D-Convolution Neural Network, Long Short-Term Memory, and Bidirectional-Long Short-Term Memory, were utilized. In comparison to other algorithms, bidirectional Long Short-Term Memory model has resulted in higher accuracy (96.24%), loss (0.2524 in value), precision (92.87%), and recall (92.87%) in comparison to other machine and DL algorithms. Further, the proposed model was deployed on the robotic structure for real-time detection for improvement of social-emotional pragmatic responses in individuals with deficits. The approach can serve as a potential tool for the individuals with pragmatic communication deficits.

PMID:40079556 | DOI:10.1177/09544119251325331

J Am Heart Assoc. 2025 Mar 13:e036946. doi: 10.1161/JAHA.124.036946. Online ahead of print.

ABSTRACT

Electronic health records (EHR) have revolutionized cardiovascular disease (CVD) research by enabling comprehensive, large-scale, and dynamic data collection. Integrating EHR data with advanced analytical methods, including artificial intelligence (AI), transforms CVD risk prediction and management methodologies. This review examines the advancements and challenges of using EHR in developing CVD prediction models, covering traditional and AI-based approaches. While EHR-based CVD risk prediction has greatly improved, moving from models that integrate real-world data on medication use and imaging, challenges persist regarding data quality, standardization across health care systems, and geographic variability. The complexity of EHR data requires sophisticated computational methods and multidisciplinary approaches for effective CVD risk modeling. AI's deep learning enhances prediction performance but faces limitations in interpretability and the need for validation and recalibration for diverse populations. The future of CVD risk prediction and management increasingly depends on using EHR and AI technologies effectively. Addressing data quality issues and overcoming limitations from retrospective data analysis are critical for improving the reliability and applicability of risk prediction models. Integrating multidimensional data, including environmental, lifestyle, social, and genomic factors, could significantly enhance risk assessment. These models require continuous validation and recalibration to ensure their adaptability to diverse populations and evolving health care environments, providing reassurance about their reliability.

PMID:40079336 | DOI:10.1161/JAHA.124.036946

Brief Bioinform. 2025 Mar 4;26(2):bbaf114. doi: 10.1093/bib/bbaf114.

ABSTRACT

An accurate deep learning predictor is needed for enzyme optimal temperature (${T}_{opt}$), which quantitatively describes how temperature affects the enzyme catalytic activity. In comparison with existing models, a new model developed in this study, Seq2Topt, reached a superior accuracy on ${T}_{opt}$ prediction just using protein sequences (RMSE = 12.26°C and R2 = 0.57), and could capture key protein regions for enzyme ${T}_{opt}$ with multi-head attention on residues. Through case studies on thermophilic enzyme selection and predicting enzyme ${T}_{opt}$ shifts caused by point mutations, Seq2Topt was demonstrated as a promising computational tool for enzyme mining and in-silico enzyme design. Additionally, accurate deep learning predictors of enzyme optimal pH (Seq2pHopt, RMSE = 0.88 and R2 = 0.42) and melting temperature (Seq2Tm, RMSE = 7.57 °C and R2 = 0.64) were developed based on the model architecture of Seq2Topt, suggesting that the development of Seq2Topt could potentially give rise to a useful prediction platform of enzymes.

PMID:40079266 | DOI:10.1093/bib/bbaf114

Radiol Cardiothorac Imaging. 2025 Apr;7(2):e240382. doi: 10.1148/ryct.240382.

ABSTRACT

Pleuroparenchymal fibroelastosis (PPFE) is an interstitial lung disease (ILD) characterized at CT by upper lobe-predominant pleural thickening and subpleural fibrosis and histologically by visceral pleural fibrosis and subpleural fibroelastosis. Although initially classified as a rare idiopathic interstitial pneumonia, many cases are related to known risk factors, particularly hematopoietic stem cell and lung transplant, or observed in association with other ILDs. This review summarizes the diagnostic criteria for PPFE and illustrates the CT and histologic manifestations, aiming to familiarize the radiologist with the range of findings suggestive of the diagnosis. Keywords: Conventional Radiography, CT, Pulmonary, Thorax, Lung, Pleura, Complications, Transplantation, Fibrosis © RSNA, 2025.

PMID:40079759 | DOI:10.1148/ryct.240382

Elife. 2025 Mar 13;13:RP97330. doi: 10.7554/eLife.97330.

ABSTRACT

Antimicrobial peptides (AMPs) are attractive candidates to combat antibiotic resistance for their capability to target biomembranes and restrict a wide range of pathogens. It is a daunting challenge to discover novel AMPs due to their sparse distributions in a vast peptide universe, especially for peptides that demonstrate potencies for both bacterial membranes and viral envelopes. Here, we establish a de novo AMP design framework by bridging a deep generative module and a graph-encoding activity regressor. The generative module learns hidden 'grammars' of AMP features and produces candidates sequentially pass antimicrobial predictor and antiviral classifiers. We discovered 16 bifunctional AMPs and experimentally validated their abilities to inhibit a spectrum of pathogens in vitro and in animal models. Notably, P076 is a highly potent bactericide with the minimal inhibitory concentration of 0.21 μM against multidrug-resistant Acinetobacter baumannii, while P002 broadly inhibits five enveloped viruses. Our study provides feasible means to uncover the sequences that simultaneously encode antimicrobial and antiviral activities, thus bolstering the function spectra of AMPs to combat a wide range of drug-resistant infections.

PMID:40079572 | DOI:10.7554/eLife.97330