Hierarchical Negative Sampling Based Graph Contrastive Learning Approach for Drug-Disease Association Prediction.

Predicting potential drug-disease associations (RDAs) plays a pivotal role in elucidating therapeutic strategies for diseases and facilitating drug repositioning, making it of paramount importance. However, existing methods are constrained and rely heavily on limited domain-specific knowledge, impeding their ability to effectively predict candidate associations between drugs and diseases. Moreover, the simplistic definition of unknown information pertaining to drug-disease relationships as negative samples presents inherent limitations. To overcome these challenges, we introduce a novel hierarchical negative sampling-based graph contrastive model, termed HSGCLRDA, which aims to forecast latent associations between drugs and diseases. In this study, HSGCLRDA integrates the association information as well as similarity between drugs, diseases and proteins. Meanwhile, the model constructs a drug-disease-protein heterogeneous network. Subsequently, employing a hierarchical structural sampling technique, we establish reliable negative drug-disease samples utilizing PageRank algorithms. Utilizing meta-path aggregation within the heterogeneous network, we derive low-dimensional representations for drugs and diseases, thereby constructing global and local feature graphs that capture their interactions comprehensively. To obtain representation information, we adopt a self-supervised graph contrastive approach that leverages graph convolutional networks (GCNs) and second-order GCNs to extract feature graph information. Furthermore, we integrate a contrastive cost function derived from the cross-entropy cost function, facilitating holistic model optimization. Experimental results obtained from benchmark datasets not only showcase the superior performance of HSGCLRDA compared to various baseline methods in predicting RDAs but also emphasize its practical utility in identifying novel potential diseases associated with existing drugs through meticulous case studies.

Predicting Potential Drug-Disease Associations Based on Hypergraph Learning with Subgraph Matching.

The search for potential drug-disease associations (DDA) can speed up drug development cycles, reduce costly wasted resources, and accelerate disease treatment by repurposing existing drugs that can control further disease progression. As technologies such as deep learning continue to mature, many researchers tend to use emerging technologies to predict potential DDA. The performance of DDA prediction is still challenging and there is some space for improvement due to issues such as the small number of existing associations and possible noise in the data. To better predict DDA, we propose a computational approach based on hypergraph learning with subgraph matching (HGDDA). In particular, HGDDA first extracts feature subgraph information in the validated drug-disease association network and proposes a negative sampling strategy based on similarity network to reduce the data imbalance. Second, the hypergraph Unet module is used by extracting Finally, the potential DDA is predicted by designing a hypergraph combination module to convolution and pooling the two constructed hypergraphs separately, and calculating the difference information between the subgraphs using cosine similarity for node matching. The performance of HGDDA is verified under two standard datasets by 10-fold cross-validation (10-CV), and the results outperform existing drug-disease prediction methods. In addition, to validate the overall utility of the model, the top 10 drugs for the specific disease are predicted through the case study and validated using the CTD database.

ISLMI:Predicting lncRNA-miRNA Interactions Based on Information Injection and Second-Order Graph Convolution Network.

Studies have shown that IncRNA-miRNA interactions can affect cellular expression at the level of gene molecules through a variety of regulatory mechanisms and have important effects on the biological activities of living organisms. Several biomolecular network-based approaches have been proposed to accelerate the identification of lncRNA-miRNA interactions. However, most of the methods cannot fully utilize the structural and topological information of the lncRNA-miRNA interaction network. In this article, we proposed a new method, ISLMI, a prediction model based on information injection and second order graph convolution network(SOGCN). The model calculated the sequence similarity and Gaussian interaction profile kernel similarity between lncRNA and miRNA, fused them to enhance the intrinsic interaction between the nodes, using SOGCN to learn second-order representations of similarity matrix information. At the same time, multiple feature representations obtain using different graph embedding methods were also injected into the second-order graph representation. Finally, matrix complementation was used to increase the model accuracy. The model combined the advantages of different methods and achieved reliable performance in 5-fold cross-validation, significantly improved the performance of predicting lncRNA-miRNA interactions. In addition, our model successfully confirmed the superiority of ISLMI by comparing it with several other model algorithm.

RLF-LPI: An ensemble learning framework using sequence information for predicting lncRNA-protein interaction based on AE-ResLSTM and fuzzy decision.

Long non-coding RNAs (lncRNAs) play a regulatory role in many biological cells, and the recognition of lncRNA-protein interactions is helpful to reveal the functional mechanism of lncRNAs. Identification of lncRNA-protein interaction by biological techniques is costly and time-consuming. Here, an ensemble learning framework, RLF-LPI is proposed, to predict lncRNA-protein interactions. The RLF-LPI of the residual LSTM autoencoder module with fusion attention mechanism can extract the potential representation of features and capture the dependencies between sequences and structures by k-mer method. Finally, the relationship between lncRNA and protein is learned through the method of fuzzy decision. The experimental results show that the ACC of RLF-LPI is 0.912 on ATH948 dataset and 0.921 on ZEA22133 dataset. Thus, it is demonstrated that our proposed method performed better in predicting lncRNA-protein interaction than other methods.

Ferroelectrically tunable magnetism in BiFeO3/BaTiO3 heterostructure revealed by the first-principles calculations.

The perovskite oxide interface has attracted extensive attention as a platform for achieving strong coupling between ferroelectricity and magnetism. In this work, robust control of magnetoelectric (ME) coupling in the BiFeO3/BaTiO3 (BFO/BTO) heterostructure (HS) was revealed by using the first-principles calculation. Switching of the ferroelectric polarization of BTO induce large ME effect with significant changes on the magnetic ordering and easy magnetization axis, making up for the weak ME coupling effect of single-phase multiferroic BFO. In addition, the Dzyaloshinskii-Moriya interaction (DMI) and the exchange coupling constants J for the BFO part of the HSs are simultaneously manipulated by the ferroelectric polarization, especially the DMI at the interface is significantly enhanced, which is three or four times larger than that of the individual BFO bulk. This work paves the way for designing new nanomagnetic devices based on the substantial interfacial ME effect.

InTeI: a novel wide-bandgap 2D material with desirable stability and highly anisotropic carrier mobility.

Recently, stable 2D wide-bandgap semiconductors with excellent electronic and photoelectronic properties have attracted much scientific and technological interest. In this study, we predict a novel InTeI monolayer which has a wide bandgap of 2.735 eV and a anisotropic electron mobility as high as 12 137.80 cm2 V-1 s-1 based on first-principles calculations. With an exfoliating energy lower than that of monolayer phosphorene, it is feasible to synthesize the 2D InTeI monolayer through mechanical exfoliation from their 3D bulk crystals. Remarkably, the monolayer InTeI achieves the indirect-to-direct bandgap transition under a small in-plane uniaxial strain, while a quasi-direct bandgap can be achieved in the InTeI nanosheets with elevated thickness. The InTeI monolayer and nanosheets have suitable band alignments in the visible-light excitation region. In addition, our theoretical simulations determine that 2D InTeI materials exhibit more excellent oxidation resistance than black phosphorene. The results not only identify a novel class of 2D wide-bandgap semiconductors but also demonstrate their potential applications in nanoelectronics and optoelectronics.

Core-shell nanostructures introduce multiple potential barriers to enhance energy filtering for the improvement of the thermoelectric properties of SnTe.

Using dispersed nanostructures to induce an energy filtering effect is an easy and effective mechanism to optimize the performance of bulk thermoelectric materials. Compared with other nanostructures, core-shell nanostructures possess more interfaces and multiple potential barriers, which would lead to a significant impact on the thermal and electrical properties of materials. In this paper, after BiCuSeO alloy doping into SnTe, SnO2 layers were formed at the interfaces and the BiCuSeO nanoparticles were wrapped in the SnO2 shell during the following high temperature solid state reaction. The formation of SnO2 layers could be observed and confirmed by X-ray diffraction (XRD) and scanning electron microscopy (SEM). BiCuSeO@SnO2 core-shell nanostructures can introduce multiple potential barriers to enhance the energy filtering effect. Once the BiCuSeO doping concentration was over 3%, the carrier concentration could decrease to about 10% while the mobility increases to 350% compared to the values of the undoped sample at room temperature. Meanwhile, the Seebeck coefficients were improved to 176.05 µV K-1 at 835 K. Additionally, due to the scattering of core-shell nanostructures for the phonons, a lower thermal conductivity is achieved with a value of 1.04 W m-1 K-1 at 835 K in Sn1.03Te-5% BiCuSeO. Combined with the improvement of thermal and electrical properties by the BiCuSeO@SnO2 core-shell, a high ZT value of ∼1.21 was achieved for Sn1.03Te-5% BiCuSeO at 835 K, which was enhanced by 190% compared to pristine SnTe.

High Thermoelectric Performance of Bi0.46Sb1.54Te3-SnTe: Synergistic Modulation of Electrical and Thermal Transport by the Introduction of Thermoelectric Hetero Nano Region.

The thermoelectric hetero nano region, as a new strategy, can effectively modulate the electrical and thermal transport properties. In this study, the thermoelectric hetero nano region is explored to improve the thermoelectric performance for Bi0.46Sb1.54Te3 material at room temperature, and a high ZT of 1.45 at 325 K has been achieved. We introduce the thermoelectric hetero nano SnTe regions in a Bi0.46Sb1.54Te3 matrix by mechanical alloying and spark plasma sintering technique, which decouples the relation between electrical and thermal transport properties. The improved electrical conductivity can be attributed to the increase in carrier concentration due to the increased point defects and Bi/SbTe antisite defects. Thermoelectric hetero nano regions effectively scatter the acoustic phonon and thus induce the low lattice thermal conductivity of 0.33 W m-1 K-1. Due to the synergistic modulation of electrical and thermal transport by the introduction of the thermoelectric hetero nano region, a high ZT value of 1.45 is realized at 325 K.

Transition metal embedded C3N monolayers as promising catalysts for the hydrogen evolution reaction.

Transition metal (TM) doped or TM, N co-doped carbon materials have attracted increasing attention as efficient catalysts for the hydrogen evolution reaction (HER), to replace Pt or reduce the usage of Pt. By using first-principles calculations, the TM-embedded C3N monolayer (TM@C3N) has been theoretically investigated for HER, for which eighteen TMs are selected from the 3d, 4d, and 5d rows. The M-CC catalysts, with the TM atom embedded into the C-C double atomic vacancy, are the most stable among the various TM@C3N materials. All the M-CC catalysts show metallic conductivity and high thermal stability. The hydrogen binding free energy for the M-CC catalysts can be optimized to be close to 0 eV by choosing a suitable TM, and the kinetic barrier under the Tafel mechanism for further gaseous hydrogen evolution can be reduced to as low as 0.58 eV. These results suggest that the HER catalytic activities of the M-CC catalysts are likely comparable or even higher than those of the well-explored MoS2 nanostructures or Pt catalysts. Moreover, the HER activities of the M-CC catalysts can be illustrated by the electronic state distribution near the Fermi level of the catalytically active sites. This study provides a new possibility for cost-efficient HER catalysts of high activity and for the application of C3N nanostructures.

Enhancement of Thermoelectric Properties in Pd-In Co-Doped SnTe and Its Phase Transition Behavior.

SnTe has attracted more and more attention due to the similar band and crystal structure with high performance thermoelectric materials PbTe. Here, we introduced Pd into SnTe and the valence band convergence was confirmed by first-principles calculation. In the experimental process, we found that Pd-doped SnTe exhibit a reduced thermal conductivity because of softening chemical bonds and grain refining effects. To further improve the thermoelectric performance, Pd-In codoped SnTe samples were prepared, and the abnormal change of thermal conductivity was observed. The results of synchrotron powder diffraction suggest that the local phase transition (local structural distortions) near 400 K results in the first turn on thermal conductivity. Similarly, the second local phase transition in near 600 K observed by neutron powder diffraction lead to a decrease thermal conductivity of the sample. Finally, a peak thermoelectric figure of merit (ZT) ≈ 1.51 has been obtained in Sn0.98Pd0.025In0.025Te at 800 K.

Seeking large Seebeck effects in LaX(X = Mn and Co)O3/SrTiO3 superlattices by exploiting high spin-polarized effects.

SrTiO3-based transition-metal oxide heterostructures with superconducting, ferromagnetic, ferroelectric, and ferroelastic properties exhibit high application potential in the fields of energy storage, energy conversion, and spintronic devices. Meanwhile, high effective (charge)-Seebeck coefficient materials composed of a ferromagnetic layer and SrTiO3 insulator layer have been achieved but we still have blocks to pursuing high spin-Seebeck coefficient materials. Here, we use first-principles calculations combined with spin-resolved Boltzmann transport theory to investigate the spin- and effective-Seebeck coefficients in the LaX(X = Mn and Co)O3/SrTiO3 superlattice. Compared with the LaMnO3/SrTiO3 superlattice, LaCoO3/SrTiO3 with ferromagnetic ordering has high spin polarization, relatively low valence valley degeneracy but high effective mass. Utilizing these characteristics, the maximum spin-Seebeck coefficient of LaMnO3/SrTiO3 is -152 µV K-1 at 450 K along the cross-plane direction, while LaCoO3/SrTiO3 reaches -247 µV K-1 under the same conditions. Interestingly, the spin- and effective-Seebeck coefficients are amazingly consistent with each other below 200 K, which indicates that one spin channel (spin-up or spin-down) dominates the carrier transport, and the other one (spin-down or spin-up) is filtered out. These characteristics are mainly associated with the magnetic MnO2/CoO2 layers with distinct dxy and dz2 orbitals near the Fermi level. Our results clarify the relationship of spin- and effective-Seebeck coefficients and indicate that SrTiO3-based transition metal oxide heterointerfaces are a key candidate for spin caloritronics.

Valence mediated tunable magnetism and electronic properties by ferroelectric polarization switching in 2D FeI2/In2Se3 van der Waals heterostructures.

Exploring two-dimensional (2D) materials with both ferromagnetic and ferroelectric properties is scientifically interesting and of great technical importance to numerous functionalities in nanoscale devices. In this work, we have demonstrated a strong magnetoelectric coupling that appeared in the 2D FeI2/In2Se3 van der Waals heterostructure. FeI2 layers undergo a transition from ferromagnetic to antiferromagnetic by reversing the direction of ferroelectric polarization. First-principles calculation predicts a new magnetoelectronic coupling mechanism which is completely different from the Dzyaloshinskii-Moriya (DM) effect in multiferroic materials. Because of the polarization discontinuity at the interface, the valence states of Fe ions change between +2 and +3 for two different polarization directions, leading to the magnetic interaction variation between the direct exchange and I ion mediated superexchange. Moreover, metallic 2D electron gas (2DEG) transfers from the surface of FeI2 to In2Se3 when the polarization reverses, which induces the spin polarization of the heterostructure varying from 93% to 0%. Our work is the first realization of manipulation magnetism by an electric field in full 2D van der Waals heterostructures.

Topological phase transition with p orbitals in the exciton-polariton honeycomb lattice.

We study the topological phase transition with the TE-TM splitting in the p-orbital exciton-polariton honeycomb lattice. We find that some Dirac points survive at the high-symmetry points with space-inversion symmetry breaking, which reflects the characteristic of p orbitals. A phase diagram is obtained by the gap Chern number, from which the topological phase transition takes place in the intermediate gap. There is no topological phase transition in the bottom or top gap, and its edge state has the potential application for transporting signals in optoelectronic devices. When taking into account the non-degenerate p orbitals, we find that the bottom gap arises owing to the competition between the Zeeman energy and rotating angular velocity, and topological phase transition also appears in the complete gaps. These results can facilitate the experimental investigations of the topological properties of p-orbital exciton-polariton lattice structure.

Tuning the magnetism of two-dimensional hematene by ferroelectric polarization.

Magnetism in two-dimensional (2D) materials, that is, a 2D version of the magnetism of three-dimensional bulk materials, and the associated novel physics have recently been the focus of many spintronics researchers. Here we investigate the manipulation of 2D magnetism at the interfaces of ferromagnetic/ferroelectric hematene/BaTiO3(001) heterostructures (HSs) fabricated via a precisely chosen sequence. By introducing four types of interfaces of 2D hematene and three-dimensional BaTiO3 that induce different oxygen environments, the control of magnetism is directly demonstrated from first-principles. An obvious 2D electron gas originates from the Fe-3d and O-2p hybridization; the electron gas is sensitive to the interfacial atomic displacements. Robust control of both the direction and magnitude of the net magnetization has been realized for an Fe/TiO2 terminated bilayer HS. The electron occupancies of the dxy and dxz orbitals and changes to the Fe-O bond play a key role in determining the magnetism of our systems. Our work not only demonstrates the technique's potential for manipulating magnetism in 2D hematene, but also sheds light on the underlying mechanism and the fundamental properties of hematene HSs.

Robust manipulation of magnetism in LaAO3/BaTiO3 (A = Fe, Mn and Cr) superstructures by ferroelectric polarization.

Robust control of magnetism is both fundamentally and practically meaningful and highly desirable, although it remains a big challenge. In this work, perovskite oxide superstructures LaFeO3/BaTiO3 (LFO/BTO), LaMnO3/BaTiO3 (LMO/BTO) and LaCrO3/BaTiO3 (LCO/BTO) (001) are designed to facilitate tuning of magnetism by the electric field from ferroelectric polarization, and are systemically investigated via first-principles calculations. The results show that the magnetic ordering, conductivity and exchange interactions can be controlled simultaneously or individually by the reorientation of the ferroelectric polarization of BTO in these designed superstructures. Self-consistent calculations within the generalized gradient approximation plus on-site Coulomb correction did not produce distinct rotations of oxygen octahedra, but there were obvious changes in bond length between oxygen and the cations. These changes cause tilting of the oxygen octahedra and lead to spin, orbital and bond reconstruction at the interface, which is the structural basis responsible for the manipulation. With the G-type antiferromagnetic (G-AFM) ordering unchanged for both ±P cases, a metal-insulator transition can be observed in the LFO/BTO superstructure, which is controlled by the LFO thin film. The LMO/BTO system has A-type antiferromagnetic (A-AFM) ordering with metallic behavior in the +P case, while it shifts to a half-metallic ferromagnetic ordering when the direction of the polarization is switched. LCO/BTO exhibits C-type antiferromagnetic (C-AFM) and G-AFM orders in the +P and -P cases, respectively. The three purpose-designed superstructures with robust intrinsic magnetoelectric coupling are a particularly interesting model system that can provide guidance for the development of this field for future applications.

Magneto-Seebeck effect in Co2FeAl/MgO/Co2FeAl: first-principles calculations.

The magneto-Seebeck effect has recently attracted considerable attention because of its novel fundamental physics and future potential application in spintronics. Herein, employing first-principles calculations and the spin-resolved Boltzmann transport theory, we have systematically investigated the electronic structures and spin-related transport properties of Co2FeAl/MgO/Co2FeAl multilayers with parallel (P) and anti-parallel (AP) magnetic alignment. Our results indicate that the sign of tunneling magneto-Seebeck (TMS) value with Co2/O termination is consistent with that of the measured experimental result although its value (-221%) at room temperature is smaller than the experimental one (-95%). The calculated spin-Seebeck coefficients of the Co2/O termination with P and AP states and the FeAl/O termination with the AP state are all larger than other typical Co2MnSi/MgO/Co2MnSi heterostructures. By analyzing the geometries, electronic structures, and magnetic behaviors of two different terminations (Co2/O and FeAl/O terminations), we find that the two terminations in the interface region form anti-bonding and bonding states, reconstructing the energy gap, changing the magnetic moment of O atoms, and improving the spin-polarization (-82%). This phenomenon can be ascribed to the charge transfer and hybridization between Co/Fe 3d and O 2p states, which also results in a bowknot orbital shape of Co atoms with Co2/O termination and an ankle shape of Co atoms with FeAl/O termination far away from the interface. Moreover, there are spin-splitting transmission gaps with the Co2/O-termination around the Fermi level, while the transmission gaps with the FeAl/O-termination are closed and thus show a typical metallic character. Our findings will guide the experimental design of magneto-Seebeck devices for future spintronic applications.

New monolayer ternary In-containing sesquichalcogenides BiInSe3, SbInSe3, BiInTe3, and SbInTe3 with high stability and extraordinary piezoelectric properties.

Looking for the high-performance alternatives to conventional lead-containing piezoelectric materials such as lead zirconate titanate (PZT) is absolutely vital for the development of low-dimensional innovative piezoelectric devices. Herein, we present our first-principles calculations on several new monolayers consisting of ternary In-containing sesquichalcogenides, which exhibit high stability and extraordinary piezoelectric properties. Our calculations predict that the in-plane (d11) and out-of-plane (d31) piezoelectric coefficients of BiInSe3, SbInSe3, BiInTe3, and SbInTe3 monolayers are much larger than those of most previously reported two-dimensional (2D) materials and widely studied wurtzite-type bulk piezoelectrics. Very strikingly, BiInTe3 monolayer possesses a d11 as high as 362 pm V-1 due to its mechanical flexibility, which is the highest among those reported in 2D materials and for the first time reaches those (∼360 pm V-1) in bulk lead-containing piezoelectric materials such as PZT. The theoretical predictions of the giant piezoelectricity in these 2D materials suggest that they have great potentials for the applications in atomically thin lead-free piezoelectric devices such as sensors and energy harvesters.

Engineering electrical transport in α-MgAgSb to realize high performances near room temperature.

α-MgAgSb shows promise as a potential new low-temperature thermoelectric (TE) material and has been widely researched recently. We explored the effects of sintering conditions on the properties of MgAgSb-based thermoelectric materials through manipulating a spark plasma sintering system (SPS), where Ag vacancies and Mg point defects play a dominant role. The transport properties of MgAgSb were optimized effectively and efficiently, especially for electrical transport. As a result, we obtained a steady power factor (PF) of ∼17 µW cm-1 K-2, owing to the optimal carrier concentration of 9.8 × 1019 cm-3. Additionally, α-MgAgSb exhibits an ultralow lattice thermal conductivity of around 0.45 Wm-1 K-1 at 375 K. More importantly, a high ZT value of 0.85 was achieved below 375 K, approaching room temperature.

Multifield Control of Domains in a Room-Temperature Multiferroic 0.85BiTi0.1Fe0.8Mg0.1O3-0.15CaTiO3 Thin Film.

Single-phase materials that combine electric polarization and magnetization are promising for applications in multifunctional sensors, information storage, spintronic devices, etc. Following the idea of a percolating network of magnetic ions (e.g., Fe) with strong superexchange interactions within a structural scaffold with a polar lattice, a solid solution thin film with perovskite structure at a morphotropic phase boundary with a high level of Fe atoms on the B site of perovskite structure is deposited to combine both ferroelectric and ferromagnetic ordering at room temperature with magnetoelectric coupling. In this work, a 0.85BiTi0.1Fe0.8Mg0.1O3-0.15CaTiO3 thin film has been deposited by pulsed laser deposition (PLD). Both the ferroelectricity and the magnetism were characterized at room temperature. Large polarization and a large piezoelectric effective coefficient d33 were obtained. Multifield coupling of the thin film has been characterized by scanning force microscopy. Ferroelectric domains and magnetic domains could be switched by magnetic field ( H), electric field ( E), mechanical force ( F), and, indicating that complex cross-coupling exists among the electric polarization, magnetic ordering and elastic deformation in 0.85BiTi0.1Fe0.8Mg0.1O3-0.15CaTiO3 thin film at room temperature. This work also shows the possibility of writing information with electric field, magnetic field, and mechanical force and then reading data by magnetic field. We expect that this work will benefit information applications.

Hf/Sb co-doping induced a high thermoelectric performance of ZrNiSn: First-principles calculation.

Previous experiments showed that Hf/Sb co-doping in ZrNiSn impressively improved the electrical conductivity (σ). To explore the physical reasons for this improvement, the electronic structures of HfxZr1-xNiSn1-ySby (x = 0, 0.25, 0.5; y = 0, 0.02) have been systematically investigated by using the first-principles method and semiclassical Boltzmann transport theory. 50% Hf doping at Zr site in ZrNiSn simultaneously increases the degeneracy and dispersion of energy bands near the conduction band edge, which are helpful to optimizing Seebeck coefficient and slightly improving σ. Furthermore, 2% Sb co-doping at Sn site in Hf0.5Zr0.5NiSn not only increases total density of states near the Fermi energy but also retains high mobility, and N v reaches eleven at the conduction band minimum, thereby inducing a large improvement in σ. Additionally, the Bader charge analysis shows the reason why Sb co-doping supplies more electrons. It is most likely derived from that Sb loses more electrons and Sb-Ni has a stronger hybridization than Sn-Ni. Moreover, we predict that the ZT of Hf0.5Zr0.5NiSn0.98Sb0.02 at 1000 K can reach 1.37 with the carrier concentration of 7.56 × 1018 cm-3, indicating that Hf/Sb co-doping may be an effective approach in optimizing thermoelectric properties of ZrNiSn alloy compounds.