Single-Image-Based Deep Learning for Precise Atomic Defect Identification.

Defect engineering is widely used to impart the desired functionalities on materials. Despite the widespread application of atomic-resolution scanning transmission electron microscopy (STEM), traditional methods for defect analysis are highly sensitive to random noise and human bias. While deep learning (DL) presents a viable alternative, it requires extensive amounts of training data with labeled ground truth. Herein, employing cycle generative adversarial networks (CycleGAN) and U-Nets, we propose a method based on a single experimental STEM image to tackle high annotation costs and image noise for defect detection. Not only atomic defects but also oxygen dopants in monolayer MoS2 are visualized. The method can be readily extended to other two-dimensional systems, as the training is based on unit-cell-level images. Therefore, our results outline novel ways to train the model with minimal data sets, offering great opportunities to fully exploit the power of DL in the materials science community.

Accurate Conformation Sampling via Protein Structural Diffusion.

Accurate sampling of protein conformations is pivotal for advances in biology and medicine. Although there has been tremendous progress in protein structure prediction in recent years due to deep learning, models that can predict the different stable conformations of proteins with high accuracy and structural validity are still lacking. Here, we introduce UFConf, a cutting-edge approach designed for robust sampling of diverse protein conformations based solely on amino acid sequences. This method transforms AlphaFold2 into a diffusion model by implementing a conformation-based diffusion process and adapting the architecture to process diffused inputs effectively. To counteract the inherent conformational bias in the Protein Data Bank, we developed a novel hierarchical reweighting protocol based on structural clustering. Our evaluations demonstrate that UFConf outperforms existing methods in terms of successful sampling and structural validity. The comparisons with long-time molecular dynamics show that UFConf can overcome the energy barrier existing in molecular dynamics simulations and perform more efficient sampling. Furthermore, We showcase UFConf's utility in drug discovery through its application in neural protein-ligand docking. In a blind test, it accurately predicted a novel protein-ligand complex, underscoring its potential to impact real-world biological research. Additionally, we present other modes of sampling using UFConf, including partial sampling with fixed motif, Langevin dynamics, and structural interpolation.

Estradiol inhibits Th17 cell differentiation through inhibition of RORγT transcription by recruiting the ERα/REA complex to estrogen response elements of the RORγT promoter.

The symptoms of vaginal candidiasis exacerbate in the second half of the menstrual cycle in premenopausal women when the serum estradiol level is elevated. Estradiol has been shown to inhibit Th17 differentiation and production of antifungal IL-17 cytokines. However, little is known about the mechanisms. In the present study, we used mouse splenocytes and found that estradiol inhibited Th17 differentiation through downregulation of Rorγt mRNA and protein expression. Estradiol activated estrogen receptor (ER)α to recruit repressor of estrogen receptor activity (REA) and form the ERα/REA complex. This complex bound to three estrogen response element (ERE) half-sites on the Rorγt promoter region to suppress Rorγt expression. Estradiol induced Rea mRNA and protein expression in mouse splenocytes. Using Rea small interfering RNA to knock down Rea expression enhanced Rorγt expression and Th17 differentiation. Alternatively, histone deacetylase 1 and 2 bound to the three ERE half-sites, independent of estradiol. Histone deacetylase inhibitor MS-275 dose- and time-dependently increased Rorγt expression and subsequently enhanced Th17 differentiation. In 15 healthy premenopausal women, high serum estradiol levels are correlated with low RORγT mRNA levels and high REA mRNA levels in the vaginal lavage. These results demonstrate that estradiol upregulates REA expression and recruits REA via ERα to the EREs on the RORγT promoter region, thus inhibiting RORγT expression and Th17 differentiation. This study suggests that the estradiol/ERα/REA axis may be a feasible target in the management of recurrent vaginal candidiasis.

Node-Aligned Graph-to-Graph: Elevating Template-free Deep Learning Approaches in Single-Step Retrosynthesis.

Single-step retrosynthesis in organic chemistry increasingly benefits from deep learning (DL) techniques in computer-aided synthesis design. While template-free DL models are flexible and promising for retrosynthesis prediction, they often ignore vital 2D molecular information and struggle with atom alignment for node generation, resulting in lower performance compared to the template-based and semi-template-based methods. To address these issues, we introduce node-aligned graph-to-graph (NAG2G), a transformer-based template-free DL model. NAG2G combines 2D molecular graphs and 3D conformations to retain comprehensive molecular details and incorporates product-reactant atom mapping through node alignment, which determines the order of the node-by-node graph outputs process in an autoregressive manner. Through rigorous benchmarking and detailed case studies, we have demonstrated that NAG2G stands out with its remarkable predictive accuracy on the expansive data sets of USPTO-50k and USPTO-FULL. Moreover, the model's practical utility is underscored by its successful prediction of synthesis pathways for multiple drug candidate molecules. This proves not only NAG2G's robustness but also its potential to revolutionize the prediction of complex chemical synthesis processes for future synthetic route design tasks.

Data-driven quantum chemical property prediction leveraging 3D conformations with Uni-Mol.

Quantum chemical (QC) property prediction is crucial for computational materials and drug design, but relies on expensive electronic structure calculations like density functional theory (DFT). Recent deep learning methods accelerate this process using 1D SMILES or 2D graphs as inputs but struggle to achieve high accuracy as most QC properties depend on refined 3D molecular equilibrium conformations. We introduce Uni-Mol+, a deep learning approach that leverages 3D conformations for accurate QC property prediction. Uni-Mol+ first generates a raw 3D conformation using RDKit then iteratively refines it towards DFT equilibrium conformation using neural networks, which is finally used to predict the QC properties. To effectively learn this conformation update process, we introduce a two-track Transformer model backbone and a novel training approach. Our benchmarking results demonstrate that the proposed Uni-Mol+ significantly improves the accuracy of QC property prediction in various datasets.

Bridging Machine Learning and Thermodynamics for Accurate pK a Prediction.

Integrating scientific principles into machine learning models to enhance their predictive performance and generalizability is a central challenge in the development of AI for Science. Herein, we introduce Uni-pK a, a novel framework that successfully incorporates thermodynamic principles into machine learning modeling, achieving high-precision predictions of acid dissociation constants (pK a), a crucial task in the rational design of drugs and catalysts, as well as a modeling challenge in computational physical chemistry for small organic molecules. Uni-pK a utilizes a comprehensive free energy model to represent molecular protonation equilibria accurately. It features a structure enumerator that reconstructs molecular configurations from pK a data, coupled with a neural network that functions as a free energy predictor, ensuring high-throughput, data-driven prediction while preserving thermodynamic consistency. Employing a pretraining-finetuning strategy with both predicted and experimental pK a data, Uni-pK a not only achieves state-of-the-art accuracy in chemoinformatics but also shows comparable precision to quantum mechanics-based methods.

A comprehensive transformer-based approach for high-accuracy gas adsorption predictions in metal-organic frameworks.

Gas separation is crucial for industrial production and environmental protection, with metal-organic frameworks (MOFs) offering a promising solution due to their tunable structural properties and chemical compositions. Traditional simulation approaches, such as molecular dynamics, are complex and computationally demanding. Although feature engineering-based machine learning methods perform better, they are susceptible to overfitting because of limited labeled data. Furthermore, these methods are typically designed for single tasks, such as predicting gas adsorption capacity under specific conditions, which restricts the utilization of comprehensive datasets including all adsorption capacities. To address these challenges, we propose Uni-MOF, an innovative framework for large-scale, three-dimensional MOF representation learning, designed for multi-purpose gas prediction. Specifically, Uni-MOF serves as a versatile gas adsorption estimator for MOF materials, employing pure three-dimensional representations learned from over 631,000 collected MOF and COF structures. Our experimental results show that Uni-MOF can automatically extract structural representations and predict adsorption capacities under various operating conditions using a single model. For simulated data, Uni-MOF exhibits remarkably high predictive accuracy across all datasets. Additionally, the values predicted by Uni-MOF correspond with the outcomes of adsorption experiments. Furthermore, Uni-MOF demonstrates considerable potential for broad applicability in predicting a wide array of other properties.

Explore the impact of hypoxia-related genes (HRGs) in Cutaneous melanoma.

BACKGROUND: Cutaneous melanoma (CM) has an overall poor prognosis due to a high rate of metastasis. This study aimed to explore the role of hypoxia-related genes (HRGs) in CM. METHODS: We first used on-negative matrix factorization consensus clustering (NMF) to cluster CM samples and preliminarily analyzed the relationship of HRGs to CM prognosis and immune cell infiltration. Subsequently, we identified prognostic-related hub genes by univariate COX regression analysis and the least absolute shrinkage and selection operator (LASSO) and constructed a prognostic model. Finally, we calculated a risk score for patients with CM and investigated the relationship between the risk score and potential surrogate markers of response to immune checkpoint inhibitors (ICIs), such as TMB, IPS values, and TIDE scores. RESULTS: Through NMF clustering, we identified high expression of HRGs as a risk factor for the prognosis of CM patients, and at the same time, increased expression of HRGs also indicated a poorer immune microenvironment. Subsequently, we identified eight gene signatures (FBP1, NDRG1, GPI, IER3, B4GALNT2, BGN, PKP1, and EDN2) by LASSO regression analysis and constructed a prognostic model. CONCLUSION: Our study identifies the prognostic significance of hypoxia-related genes in melanoma and shows a novel eight-gene signature to predict the potential efficacy of ICIs.

Cucurbitacin E inhibits cellular proliferation and induces apoptosis in melanoma by suppressing HSDL2 expression.

BACKGROUND: Melanoma is among the most aggressive types of skin malignancy and can have an unpredictable clinical course. Exploration of novel therapeutic targets and their regulators remains essential for the prevention and treatment of melanoma. METHODS: HSDL2 protein levels were examined by immunohistochemistry. The roles of HSDL2 in cell proliferation and apoptosis were identified by CCK-8 and colony formation assays. The function of HSDL2 in cell apoptosis was analysed by flow cytometry. Western blotting, cell proliferation and apoptosis and a xenograft tumour model were utilized to explore the inhibitory functions and mechanisms of CuE in melanoma. RESULTS: HSDL2 is overexpressed in melanoma and promotes melanoma progression by activating the ERK and AKT pathways. CuE could inhibit the ERK and AKT pathways by decreasing HSDL2 expression; therefore, CuE could inhibit melanoma growth in vitro and in vivo. CONCLUSION: HSDL2 may be a promising therapeutic target against melanoma, and CuE can inhibit melanoma by downregulating HSDL2 expression.

Deep Subdomain Adaptation Network for Image Classification.

For a target task where the labeled data are unavailable, domain adaptation can transfer a learner from a different source domain. Previous deep domain adaptation methods mainly learn a global domain shift, i.e., align the global source and target distributions without considering the relationships between two subdomains within the same category of different domains, leading to unsatisfying transfer learning performance without capturing the fine-grained information. Recently, more and more researchers pay attention to subdomain adaptation that focuses on accurately aligning the distributions of the relevant subdomains. However, most of them are adversarial methods that contain several loss functions and converge slowly. Based on this, we present a deep subdomain adaptation network (DSAN) that learns a transfer network by aligning the relevant subdomain distributions of domain-specific layer activations across different domains based on a local maximum mean discrepancy (LMMD). Our DSAN is very simple but effective, which does not need adversarial training and converges fast. The adaptation can be achieved easily with most feedforward network models by extending them with LMMD loss, which can be trained efficiently via backpropagation. Experiments demonstrate that DSAN can achieve remarkable results on both object recognition tasks and digit classification tasks. Our code will be available at https://github.com/easezyc/deep-transfer-learning.