PGAT-ABPp: harnessing protein language models and graph attention networks for antibacterial peptide identification with remarkable accuracy.

MOTIVATION: The emergence of drug-resistant pathogens represents a formidable challenge to global health. Using computational methods to identify the antibacterial peptides (ABPs), an alternative antimicrobial agent, has demonstrated advantages in further drug design studies. Most of the current approaches, however, rely on handcrafted features and underutilize structural information, which may affect prediction performance. RESULTS: To present an ultra-accurate model for ABP identification, we propose a novel deep learning approach, PGAT-ABPp. PGAT-ABPp leverages structures predicted by AlphaFold2 and a pretrained protein language model, ProtT5-XL-U50 (ProtT5), to construct graphs. Then the graph attention network (GAT) is adopted to learn global discriminative features from the graphs. PGAT-ABPp outperforms the other fourteen state-of-the-art models in terms of accuracy, F1-score and Matthews Correlation Coefficient on the independent test dataset. The results show that ProtT5 has significant advantages in the identification of ABPs and the introduction of spatial information further improves the prediction performance of the model. The interpretability analysis of key residues in known active ABPs further underscores the superiority of PGAT-ABPp. AVAILABILITY AND IMPLEMENTATION: The datasets and source codes for the PGAT-ABPp model are available at https://github.com/moonseter/PGAT-ABPp/.

Characterization of a monkey model with experimental retinal damage induced by N-methyl-D-aspartate.

N-methyl-D-aspartate (NMDA)-induced retinal damage has been well studied in rodents, but the detailed mechanisms have not yet been characterized in nonhuman primates. Here, we characterized the retinal degenerative effects of NMDA on rhesus monkeys in vivo. NMDA saline or saline-only control was injected intravitreally to the randomly assigned eyes and contralateral eyes of four rhesus monkeys, respectively. The structural and functional changes of retina were characterized by optical coherence tomography and electroretinography on days 0, 4, 30 and 60 post injection. Both optic discs and macular areas of the NMDA-injected eyes initially presented with a transient retinal thickening, followed by continued retinal thinning. The initial, transient retinal thickening has also been observed in glaucoma patients, but this has not been reported in rodent NMDA models. This initial response was followed by loss of retina ganglion cells (RGCs), which is similar to glaucomatous optic neuropathy and other RGC-related retinal degenerations. The amplitudes of both the photopic negative response and pattern electroretinogram decreased significantly and remained low until the end of the study. Thus, the NMDA monkey model may serve as a more clinically relevant animal model of retinal damage.

AttenGpKa: A Universal Predictor of Solvation Acidity Using Graph Neural Network and Molecular Topology.

Rapid and accurate calculation of acid dissociation constant (pKa) is crucial for designing chemical synthesis routes, optimizing catalysts, and predicting chemical behavior. Despite recent progress in machine learning, predicting solvation acidity, especially in nonaqueous solvents, remains challenging due to limited experimental data. This challenge arises from treating experimental values in different solvents as distinct data domains and modeling them separately. In this work, we treat both the solutes and solvents equally from a perspective of molecular topology and propose a highly universal framework called AttenGpKa for predicting solvation acidity. AttenGpKa is trained using 26,522 experimental pKa values from 60 pure and mixed solvents in the iBonD database. As a result, our model can simultaneously predict the pKa values of a compound in various solvents, including pure water, pure nonaqueous, and mixed solvents. AttenGpKa achieves universality by using graph neural networks and attention mechanisms to learn complex effects within solute and solvent molecules. Furthermore, encodings of both solute and solvent molecules are adaptively fused to simulate the influence of the solvent on acid dissociation. AttenGpKa demonstrates robust generalization in extensive validations. The interpretability studies further indicate that our model has effectively learnt electronic and solvent effects. A free-to-use software is provided to facilitate the use of AttenGpKa for pKa prediction.

LC-MS and GC-MS based metabolomics analysis revealed the impact of tea trichomes on the chemical and flavor characteristics of white tea.

To explore the influence of tea trichomes on the quality of white tea, liquid chromatography quadrupole time-of-flight mass spectrometry (LC-Q-TOF-MS), and headspace solid phase microextraction gas chromatography-mass spectrometry (HS-SPME-GC-MS) were used to identify non-volatile and volatile compounds white tea without trichomes (WTwt) and pure trichomes (PT). It was found that the bitter and astringent compounds, caffeine (CAF), epigallocatechin gallate (EGCG), epicatechin gallate (ECG) and flavonol glycosides, were mainly enriched in the WTwt, with 16.3-fold, 47.1-fold and 28.7-fold decrease in CAF and EGCG and ECG, respectively, and the content of these compounds in PT were lower than the taste thresholds. In PT, kaempferol-3-O-(p-coumaroyl)-glucoside and kaempferol-3-O-(di-p-coumaroyl)-glucoside were non-volatile marker compounds, and decanal was significant aroma contributor with rOAV = 250.86. Moreover, the compounds in trichomes mainly contributed to the fruity and floral aroma of white tea, among which benzyl alcohol, (E)-geranylacetone, decanal, dodecanal and 6-methyl-5-hepten-2-one were the crucial aroma components, which were 2.1, 1.7, 1.8, 1.4 and 2.2 times as much as the WTwt in the PT, respectively. In conclusion, trichomes can improve the quality of white tea by reducing the bitterness and astringency, increasing the umami, as well as enhancing the fruity and floral aromas.

Perioperative use of immersive head-mounted virtual reality display enhances patient satisfaction in great saphenous vein surgery: a single-center clinical observation study.

OBJECTIVE: To explore the application effect of head-mounted virtual reality display immersive experience in improving the perioperative satisfaction of patients undergoing great saphenous vein surgery. METHODS: A total of 158 patients undergoing saphenous vein surgery at the First Affiliated Hospital, Jiangxi Medical College, Nanchang University from January 2020 to January 2023 were randomly divided into an observation group and a control group in a 1:1 ratio, with 79 cases in each group. The observation group received head-mounted display virtual reality immersive experience, whereas the control group received midazolam. The study compared the perioperative satisfaction, changes in preoperative and postoperative anxiety and depression scores, intraoperative blood pressure and heart rate, postoperative visual analog scale (VAS) score, and the incidence of postoperative nausea and vomiting between the two groups. Additionally, the satisfaction of patients, anesthesiologists, and chief surgeons was compared. RESULTS: All surgeries were completed successfully. Patients in the observation group exhibited higher perioperative satisfaction compared to those in the control group (P<0.001). There were no significant differences in anxiety or depression scores between the two groups before surgery (P>0.05). However, both groups showed a reduction in anxiety and depression scores postoperatively, with the observation group demonstrating lower scores than the control group (both P<0.05). The observation group also had lower intraoperative blood pressure, heart rate, postoperative VAS scores, and incidence of nausea and vomiting compared to the control group (all P<0.05). Furthermore, the satisfaction levels of the anesthesiologists and chief surgeons were higher in the observation group than in the control group (P=0.043, 0.012). CONCLUSION: Head-mounted display virtual reality immersive experience can enhance perioperative satisfaction among patients undergoing great saphenous vein surgery, reduce anxiety and depression scores, and contribute to the stabilization of hemodynamics during surgery, thereby decreasing postoperative nausea and vomiting.

Fatigue Life Data Fusion Method of Different Stress Ratios Based on Strain Energy Density.

To accurately evaluate the probabilistic characteristics of the fatigue properties of materials with small sample data under different stress ratios, a data fusion method for torsional fatigue life under different stress ratios is proposed based on the energy method. A finite element numerical modeling method is used to calculate the fatigue strain energy density during fatigue damage. Torsional fatigue tests under different stresses and stress ratios are carried out to obtain a database for research. Based on the test data, the Wt-Nf curves under a single stress ratio and different stress ratios are calculated. The reliability of the models is illustrated by the scatter band diagram. More than 85% of points are within ±2 scatter bands, indicating that the fatigue life under different stress ratios can be represented by the same Wt-Nf curve. Furthermore, P-Wt-Nf prediction models are established to consider the probability characteristics. According to the homogeneity of the Wt-Nf model under different stress ratios, we can fuse the fatigue life data under different stress ratios and different strain energy densities. This data fusion method can expand the small sample test data and reduce the dispersion of the test data between different stress ratios. Compared with the pre-fusion data, the standard deviations of the post-fusion data are reduced by a maximum of 21.5% for the smooth specimens and 38.5% for the notched specimens. And more accurate P-Wt-Nf curves can be obtained to respond to the probabilistic properties of the data.

An insight into trichomes-deficiency and trichomes-rich black teas by comparative metabolomics: The impact of oxidized trichomes on metabolic profiles and infusion color.

Tea trichomes were regarded as an essential evaluation index for reflecting tea flavor quality in terms of aroma and influence on infusion color. This study reveals the impact of golden oxidized trichomes on the color, volatile and non-volatile metabolites of black teas through comparative metabolomics combined quantitative analysis on hongbiluo (trichomes-deficiency black teas), hongjinluo (trichomes-rich black teas), and trichomes (from hongjinluo). Forty-six volatile components were detected using headspace solid-phase microextraction gas chromatography-mass spectrometry, while the results suggested that the contribution of trichomes to black teas is limited. A total of 60 marker non-volatile compounds were identified, including catechins, catechin oxidation products, flavonoid glycosides, organic acids, hydrolysable tannins and amino acids. Notably, p-coumaroyl-kaempferol glucosides, and catechin dimers demonstrated high levels in independent trichomes and showed a positive correlation with the brightness and yellow hue of black tea infusions, specifically kaempferol 3-O-di-(p-coumaroyl)-hexoside. Furthermore, results from fractional extraction analysis of separated trichomes provided that N-ethyl-2-pyrrolidinone-substituted epicatechin gallates, acylated kaempferol glycosides, and chromogenic catechins dimers, such as theaflavins, were primary color contributors in oxidized trichomes. Especially, we found that epicatechin gallate (ECG) and its derivates, 3'-O-methyl-ECG and N-ethyl-2-pyrrolidinone-substituted ECG, highly accumulated in trichomes, which may be associated with the varieties of hongbiluo and hongjinluo black teas. Eventually, addition tests were applied to verify the color contribution of trichome mixtures. Our findings employed comprehensive information revealing that golden oxidized trichomes contributed significantly to the brightness and yellow hue of black tea infusion, but their contribution to the aroma and metabolic profile is limited. These findings may contribute to the effective modulation of the infusion color during black tea production by regulating the proportion of tea trichomes or screening trichomes-rich or deficiency varieties.

Novel standardized indexes of brainstem auditory evoked potentials for predicting hearing preservation in vestibular schwannomas.

Hearing preservation (HP) during vestibular schwannomas (VSs) surgery poses a significant challenge. Although brainstem auditory evoked potentials (BAEPs) on the affected side are commonly employed to monitor cochlear nerve function, their low signal-to-noise ratio (SNR) renders them susceptible to interferences, compromising their reliability. We retrospectively analyzed the data of patients who underwent tumor resection, while binaural brainstem auditory evoked potentials (BAEPs) were simultaneously recorded during surgery. To standardize BAEPs on the affected side, we incorporated the synchronous healthy side as a reference (interval between affected and healthy side ≤ 3 min). A total of 127 patients were enrolled. Comparison of the raw BAEPs data pre- and post-tumor resection revealed that neither V-wave amplitude (Am-V) nor latency (La-V) could serve as reliable predictors of HP simultaneously. However, following standardization, V-wave latency (STIAS-La-V) and amplitude (STIAS-Am-V) emerged as stable predictors of HP. Furthermore, the intraoperative difference in V-wave amplitude (D-Am-V) predicted postoperative HP in patients with preoperative HP and remained predictive after standardization. The utilization of intraoperative synchronous healthy side BAEPs as a reference to eliminate interferences proves to be an effective approach in enhancing the reliability of BAEPs for predicting HP in VSs patients.

The role of the nucleus basalis of Meynert in neuromodulation therapy: a systematic review from the perspective of neural network oscillations.

As a crucial component of the cerebral cholinergic system and the Papez circuit in the basal forebrain, dysfunction of the nucleus basalis of Meynert (NBM) is associated with various neurodegenerative disorders. However, no drugs, including existing cholinesterase inhibitors, have been shown to reverse this dysfunction. Due to advancements in neuromodulation technology, researchers are exploring the use of deep brain stimulation (DBS) therapy targeting the NBM (NBM-DBS) to treat mental and neurological disorders as well as the related mechanisms. Herein, we provided an update on the research progress on cognition-related neural network oscillations and complex anatomical and projective relationships between the NBM and other cognitive structures and circuits. Furthermore, we reviewed previous animal studies of NBM lesions, NBM-DBS models, and clinical case studies to summarize the important functions of the NBM in neuromodulation. In addition to elucidating the mechanism of the NBM neural network, future research should focus on to other types of neurons in the NBM, despite the fact that cholinergic neurons are still the key target for cell type-specific activation by DBS.

Detoxification of tetracycline and synthetic dyes by a newly characterized Lentinula edodes laccase, and safety assessment using proteomic analysis.

Fungal laccase has strong ability in detoxification of many environmental contaminants. A putative laccase gene, LeLac12, from Lentinula edodes was screened by secretome approach. LeLac12 was heterogeneously expressed and purified to characterize its enzymatic properties to evaluate its potential use in bioremediation. This study showed that the extracellular fungal laccase from L. edodes could effectively degrade tetracycline (TET) and the synthetic dye Acid Green 25 (AG). The growth inhibition of Escherichia coli and Bacillus subtilis by TET revealed that the antimicrobial activity was significantly reduced after treatment with the laccase-HBT system. 16 transformation products of TET were identified by UPLC-MS-TOF during the laccase-HBT oxidation process. Gas chromatography-mass spectrometry (GC-MS) analysis revealed that LeLac12 could completely mineralize ring-cleavage products. LeLac12 completely catalyzed 50 mg/L TET within 4 h by adding AG (200 mg/L), while the degradation of AG was above 96% even in the co-contamination system. Proteomic analysis revealed that central carbon metabolism, energy metabolism, and DNA replication/repair were affected by TET treatment and the latter system could contribute to the formation of multidrug-resistant strains. The results demonstrate that LeLac12 is an efficient and environmentally method for the removal of antibiotics and dyes in the complex polluted wastewater.

Identification of core genes involved in the response of Apocynum venetum to salt stress based on transcriptome sequencing and WGCNA.

Apocynum venetum L. belongs to the Apocynaceae family and is a plant that is highly resistant to stress. It is important in the fields of ecology, feeding, industry and medicine. The molecular mechanism underlying salt tolerance has not been elucidated. In this study, RNA-seq based transcriptome sequencing of A. venetum leaves after 0, 2, 6, 12, 24 and 48 h of treatment with 300 mM NaCl was performed. We conducted a comprehensive analysis of the transcriptome expression profiles of A. venetum under salt stress using the WGCNA method and identified red, black, and brown as the core modules regulating the salt tolerance of A. venetum. A co-expression regulatory network was constructed to identify the core genes in the module according to the correlations between genes. The genes TRINITY_DN102_c0_g1 (serine carboxypeptidase), TRINITY_DN3073_c0_g1 (SOS signaling pathway) and TRINITY_DN6732_c0_g1 (heat shock transcription factor) in the red module were determined to be the core genes. Two core genes in the black module, TRINITY_DN9926_c0_g1 and TRINITY_DN7962_c0_g1, are pioneer candidate salt tolerance-associated genes in A. venetum. The genes in the brown module were mainly enriched in two pathways, namely photosynthesis and osmotic balance. Among them, the TRINITY_DN6321_c0_g2 and TRINITY_DN244_c0_g1 genes encode aquaporin, which is helpful for maintaining the cell water balance and plays a protective role in defending A. venetum under abiotic stress. Our findings contribute to the identification of core genes involved in the response of A. venetum to salt stress.

Identification and Validation of Basement Membrane Related LncRNA Signatures as a Novel Prognostic Model for Hepatocellular Carcinoma.

Hepatocellular carcinoma (HCC) is a significant cancer with limited treatments and a poor prognosis, with the basement membrane (BM) playing a crucial role in its initiation and growth. This study utilized data from The Cancer Genome Atlas and the Gene Expression Omnibus (GEO) databases to identify basement membrane-related genes differentially expressed in HCC. Through gene co-expression analysis, BM-associated long non-coding RNAs (lncRNAs) were discovered. LncRNAs related to HCC survival were selected via univariate analysis, and a prognostic model was constructed using LASSO regression and multivariate analysis. This model effectively classified HCC patients into high and low-risk groups, uncovering significant differences in prognosis, immune response, mutation, and drug sensitivity. Six BM-related lncRNAs (GSEC, MIR4435-2HG, AC092614.1, AC127521.1, LINC02580, and AC008050.1) were validated in normal and HCC cell lines, and the key role of AC092614.1 in regulating proliferation, migration, and invasion of HCC cells in vitro was explored. This research emphasizes the prognostic and therapeutic relevance of BM-related lncRNAs in HCC, highlighting AC092614.1's role in disease progression and as a potential target for targeted therapy.

Weakened tanning ability is an important mechanism for evolutionary skin lightening in East Asians.

The evolution of light-skin pigmentation among Eurasians is considered as an adaptation to the high-latitude environments. East Asians are ideal populations for studying skin color evolution because of the complex environment of East Asia. Here, we report a strong selection signal for the pigmentation gene phenylalanine hydroxylase (PAH) in light-skinned Han Chinese individuals. The intron mutation rs10778203 in PAH is enriched in East Asians and is significantly associated with skin color of the back of the hand in Han Chinese males (P < 0.05). In vitro luciferase and transcription factor binding assays show that the ancestral allele of rs10778203 could bind to SMAD2 and has a significant enhancer activity for PAH. However, the derived T allele (the major allele in East Asians) of rs10778203 decreases the binding activity of transcription factors and enhancer activity. Meanwhile, the derived T allele of rs10778203 shows a weaker ultraviolet radiation response in A375 cells and zebrafish embryos. Furthermore, rs10778203 decreases melanin production in transgenic zebrafish embryos after ultraviolet B (UVB) treatment. Collectively, PAH is a potential pigmentation gene that regulates skin tanning ability. Natural selection has enriched the adaptive allele, resulting in weakened tanning ability in East Asians, suggesting a unique genetic mechanism for evolutionary skin lightening in East Asians.

Unveiling cell subpopulations in T1D mouse islets using single-cell RNA sequencing.

Type 1 diabetes (T1D) is an autoimmune disease characterized by the destruction of beta cells by immune cells. The interactions among cells within the islets may be closely linked to the pathogenesis of T1D. In this study, we used single-cell RNA sequencing (scRNA-Seq) to analyze the cellular heterogeneity within the islets of a T1D mouse model. We established a T1D mouse model induced by streptozotocin and identified cell subpopulations using scRNA-Seq technology. Our results revealed 11 major cell types in the pancreatic islets of T1D mice, with heterogeneity observed in the alpha and beta cell subgroups, which may play a crucial role in the progression of T1D. Flow cytometry further confirmed a mature alpha and beta cell reduction in T1D mice. Overall, our scRNA-Seq analysis provided insights into the cellular heterogeneity of T1D islet tissue and highlighted the potential importance of alpha and beta cells in developing T1D.NEW & NOTEWORTHY In this study, we created a comprehensive single-cell atlas of pancreatic islets in a T1D mouse model using scRNA-Seq and identified 11 major cell types in the islets, highlighting the role of alpha and beta cells in T1D. This study revealed a significant reduction in the maturity alpha and beta cells in T1D mice through flow cytometry. It also demonstrated the heterogeneity of alpha and beta cells, potentially crucial for T1D progression. Overall, our scRNA-Seq analysis provided new insights for understanding and treating T1D by studying cell subtype changes and functions.

Stop codon variant in EFEMP1 is associated with primary open-angle glaucoma due to impaired regulation of aqueous humor outflow.

It is known that the actin cytoskeleton and its associated cellular interactions in the trabecular meshwork (TM) and juxtacanalicular tissues mainly contribute to the formation of resistance to aqueous outflow of the eye. Fibulin-3, encoded by EFEMP1 gene, has a role in extracellular matrix (ECM) modulation, and interacts with enzymatic ECM regulators, but the effects of fibulin-3 on TM cells has not been explored. Here, we report a stop codon variant (c.T1480C, p.X494Q) of EFEMP1 that co-segregates with primary open angle glaucoma (POAG) in a Chinese pedigree. In the human TM cells, overexpression of wild-type fibulin-3 reduced intracellular actin stress fibers formation and the extracellular fibronectin levels by inhibiting Rho/ROCK signaling. TGFß1 up-regulated fibulin-3 protein levels in human TM cells by activating Rho/ROCK signaling. In rat eyes, overexpression of wild-type fibulin-3 decreased the intraocular pressure and the fibronectin expression of TM, however, overexpression of mutant fibulin-3 (c.T1480C, p.X494Q) showed opposite effects in cells and rat eyes. Taken together, the EFEMP1 variant may impair the regulatory capacity of fibulin-3 which has a role for modulating the cell contractile activity and ECM synthesis in TM cells, and in turn may maintain normal resistance of aqueous humor outflow. This study contributes to the understanding of the important role of fibulin-3 in TM pathophysiology and provides a new possible POAG therapeutic approach.

Analyzing Molecular Dynamics Trajectories Thermodynamically through Artificial Intelligence.

Molecular dynamics simulations produce trajectories that correspond to vast amounts of structure when exploring biochemical processes. Extracting valuable information, e.g., important intermediate states and collective variables (CVs) that describe the major movement modes, from molecular trajectories to understand the underlying mechanisms of biological processes presents a significant challenge. To achieve this goal, we introduce a deep learning approach, coined DIKI (deep identification of key intermediates), to determine low-dimensional CVs distinguishing key intermediate conformations without a-priori assumptions. DIKI dynamically plans the distribution of latent space and groups together similar conformations within the same cluster. Moreover, by incorporating two user-defined parameters, namely, coarse focus knob and fine focus knob, to help identify conformations with low free energy and differentiate the subtle distinctions among these conformations, resolution-tunable clustering was achieved. Furthermore, the integration of DIKI with a path-finding algorithm contributes to the identification of crucial intermediates along the lowest free-energy pathway. We postulate that DIKI is a robust and flexible tool that can find widespread applications in the analysis of complex biochemical processes.

A Sensitive Probe of Meso-Cyanophenyl Substituted BODIPY Derivative as Fluorescent Chemosensor for the Detection of Multiple Heavy Metal Ions.

A fluorescent turn-on chemosensor (BA) was constructed by attaching bis(pyridin-2-ylmethyl)-amine (DPA) unit to the BODIPY scaffold. It can give a prominent green/yellow fluorescent response selectivity with each of Zn2+/Hg2+/Cd2+/Ca2+/Mn2+/Pb2+/Al3+. The 1:1 stoichiometry of BA and metal ions was drawn from the analysis of Job's plot. The limit detection of BA in recognition of Zn2+/Hg2+/Cd2+/Ca2+/Mn2+/Pb2+/Al3+ is ranged in 50.8-146.6 nM. There exists a linear relationship between the fluorescence intensity and concentration of metal ions (Zn2+: 4-15 µM). The mechanism of fluorescence signal "turn-on" is based on the photo induced transfer (PET) in the excited state of BA. The coordinated metal ions significantly weakened the electron-donating ability nitrogen atom in DPA, thus recovering the emission character of BODIPY. The substituted group at the phenyl ring in meso-position of BODIPY scaffold determines the recognizable list of metal ions.

Explainable Graph Neural Networks with Data Augmentation for Predicting pKa of C-H Acids.

The pKa of C-H acids is an important parameter in the fields of organic synthesis, drug discovery, and materials science. However, the prediction of pKa is still a great challenge due to the limit of experimental data and the lack of chemical insight. Here, a new model for predicting the pKa values of C-H acids is proposed on the basis of graph neural networks (GNNs) and data augmentation. A message passing unit (MPU) was used to extract the topological and target-related information from the molecular graph data, and a readout layer was utilized to retrieve the information on the ionization site C atom. The retrieved information then was adopted to predict pKa by a fully connected network. Furthermore, to increase the diversity of the training data, a knowledge-infused data augmentation technique was established by replacing the H atoms in a molecule with substituents exhibiting different electronic effects. The MPU was pretrained with the augmented data. The efficacy of data augmentation was confirmed by visualizing the distribution of compounds with different substituents and by classifying compounds. The explainability of the model was studied by examining the change of pKa values when a specific atom was masked. This explainability was used to identify the key substituents for pKa. The model was evaluated on two data sets from the iBonD database. Dataset1 includes the experimental pKa values of C-H acids measured in DMSO, while dataset2 comprises the pKa values measured in water. The results show that the knowledge-infused data augmentation technique greatly improves the predictive accuracy of the model, especially when the number of samples is small.

Interpretable Perturbator for Variable Selection in near-Infrared Spectral Analysis.

A perturbator was developed for variable selection in near-infrared (NIR) spectral analysis based on the perturbation strategy in deep learning for developing interpretation methods. A deep learning predictor was first constructed to predict the targets from the spectra in the training set. Then, taking the output of the predictor as a reference, the perturbator was trained to derive the perturbation-positive (P+) and perturbation-negative (P-) features from the spectra. Therefore, the weight (σ) of the perturbator layer can be a criterion to evaluate the importance of the variables in the spectra. Ranking the spectral variables by the criterion, the number of the variables used in the quantitative model can be obtained through cross-validation. Three NIR data sets were used to evaluate the proposed method. The root mean squared error was found to be comparable with or superior to that obtained by the commonly used methods. Moreover, the selected spectral variables are interpretable in identifying the key spectral features related to the prediction target. Therefore, the proposed method provides not only an effective tool for optimizing quantitative model, but also an efficient way for explaining spectra of multicomponent samples.

A Tailored Interface Design for Anode-Free Solid-State Batteries.

Anode-free solid-state batteries (AFSSBs) are considered to be one of the most promising high-safety and high-energy storage systems. However, low Coulombic efficiency stemming from severe deterioration on solid electrolyte/current collector (Cu foil) interface and undesirable Li dendrite growth impede their practical application, especially when rigid garnet electrolyte is used. Here, an interfacial engineering strategy between garnet electrolyte and Cu foil is introduced for stable and highly efficient AFSSBs. By utilizing the high Li ion conductivity of LiC6 layer, interfacial self-adaption ability arising from ductile lithiated polyacrylic acid polymer layer and regulated Li deposition via Li-Ag alloying reaction, the garnet-based AFSSB delivers a stable long-term operation. Additionally, when combined with a commercial LiCoO2 cathode (3.1 mAh cm-2 ), the cell also exhibits an outstanding capacity retention due to the tailored interface design. The strategies for novel AFSSBs architecture thus offer an alternative route to design next-generation batteries with high safety and high density.