Abstract: The present disclosure belongs to the field of computational biology and protein engineering technology. The present disclosure provides a protein engineering and directed evolution method based on graph deep learning and applications thereof, the method includes the following steps: S1, construction of a protein structural dataset; S2, protein graph representation; S3, graph neural network model architecture; S4, model training and performance evaluation; S5, model inference, and finally identification of potential mutations that can improve the fitness. The present disclosure can realize zero-shot, low-cost, high-efficiency, and accurate prediction of protein variants with improved properties; meanwhile, TadA8ePro with improved A-to-G base editing efficiency, Cas9Plus with higher gene knockout efficiency, and OsPHR2 transcription factor with improved binding activity are also provided.

Abstract: Disclosed herein are a structure for disassembling and assembling a cutting member and a lawn mower. The structure for disassembling and assembling the cutting member includes a limiting member, a sleeve, and a locking assembly, where the locking assembly includes a locked state and an unlocked state; when the locking assembly is in the locked state, the limiting member is locked onto the sleeve to restrict a movement stroke of the cutting member; and when the locking assembly is in the unlocked state, the limiting member can disengage from the sleeve, and the cutting member extends along a first straight line and is rotated until the first straight line forms a preset angle with a first direction to disassemble the cutting member.

Abstract: The present disclosure provides a parabolic tooth trace gear mechanism with a combined tooth profile of an end face circular arc and a parabola, which belongs to the field of gear transmission. The parabolic tooth trace gear mechanism with the combined tooth profile of the end face circular arc and the parabola includes a small wheel and a big wheel geared externally with parallel axes, tooth surfaces of the small wheel and the big wheel both have a parabolic tooth trace structure, and both are formed by the motion of a combined tooth profile curve of an end face along a contact point, contact lines are all parabolas after being unfolded along a cylindrical surface, and the parabola is determined by a meshing line parameter equation and basic design parameters such as a coincidence degree, the number of teeth, a transmission ratio and the like. When correctly installed, at least one pair of gear teeth of the small wheel and the big wheel, implement pure rolling meshing contact at a node.

Abstract: A method includes identifying, by a processing device of a content sharing platform, a request to load a content feed for a user of the content sharing platform. Based on the one or more features and using an artificial intelligence (AI) model, a content feed pace for the user is identified. A set of media items is selected based on the content feed pace. The set of media items is provided, via the content feed, for user consumption.

Abstract: A rolling gear rack mechanism with tooth profile having hyperbolic tooth line structure based on a parabolic function includes a gear and a rack. An end face tooth profile of the gear and an end face tooth profile of the rack are composed of an end face working tooth profile curve and a tooth root transition curve, and the end face tooth profile of the gear and the end face tooth profile of rack are both symmetrical left and right. The end face working tooth profile curve of the gear and the end face working tooth profile curve of the rack are parabolic, and a tooth surface of the gear and a tooth surface of the rack have a hyperbolic tooth line structure. At least one pair of gear teeth meshing points of the gear and the rack are located at nodes to achieve rolling meshing contact.

Abstract: A pre-oxidized steel sheet for press hardening, a press hardened, pre-oxidized steel sheet for a vehicle component, and a method of forming a pre-oxidized steel sheet. The pre-oxidized steel sheet includes a surface, a bulk region beneath the surface, and an oxide layer disposed on the surface. The oxide layer exhibits a thickness in the range of 20 nanometers to 60 nanometers and includes at least one of a manganese oxide, an iron oxide, a manganese iron oxide, a chromium oxide, and an amorphous silicon oxide.

Abstract: An integrated method for supporting complex system design and optimization selects first graphical-textual model frameworks in a problem domain; implements functional requirement verification in a validation domain and guides a physical architecture model; selects second graphical-textual model frameworks in a solution domain to establish a physical architecture model; generates an integrated simulation model oriented towards functional and non-functional requirements in the validation domain; selects third graphical-textual model frameworks in an optimization domain to establish a multidisciplinary optimization model, and establishes binding with preset subsystem parameters in the physical architecture model of the solution domain and output values of variables related to an objective function; automatically feeds an optimal optimization scheme obtained through iterative solving based on intelligent optimization algorithms to a system design model for automatic verification.

Abstract: A method of forming a semiconductor memory structure is disclosed. A substrate having thereon a device cell region and a contact forming region in proximity to the device cell region is provided. A memory cell transistor is formed on the substrate within the device cell region. The memory cell transistor includes a gate and a charge storage structure between the gate and the substrate. The gate comprises an extended portion within the contact forming region. A first spacer is formed on a sidewall of the gate within the device cell region. The first spacer has a first spacer height. A second spacer is formed on a sidewall of the extended portion of the gate within the contact forming region. The second spacer has a second spacer height that is greater than the first spacer height.

Abstract: A quick connector includes: a tubular connector body defining an insertion axis and having a sealing zone in the form of an inner annular groove for receiving a male element; and an axial retaining element. The inner annular groove includes a bottom and a side wall which are defined by an inner shoulder of the main tubular connector body. The inner shoulder is not contained in a transverse plane, perpendicular to the insertion axis. The connector includes, between the axial retaining element and the bottom, at least two seals and a spacer located between the seals.

Abstract: The present disclosure relates to the technical field of molecular biology, and in particular to an adeno-associated virus structural plasmid capable of improving adeno-associated virus titer, which is provided with a Rep gene expression cassette and a Cap gene expression cassette in sequence in the gene expression direction. The Rep gene expression cassette includes a rep gene regulated and transcribed by a first promoter, and a transcription termination signal fragment is arranged behind the rep gene; the Cap gene expression cassette includes a cap gene regulated and transcribed by a second promoter, wherein the first promoter includes any one of a P5 promoter, an RSV promoter, an MMTV promoter, a UBC promoter, and a U6 promoter, and the second promoter includes one or more of a CMV promoter, a CBh promoter, a CAG promoter, an EF1? promoter, and an SFFV promoter.

Abstract: Methods and devices for monitoring and maintaining a health state of an electrical device. The methods can include obtaining a first operation state of the electrical device; rendering a holographic image of the electrical device and a first health state of the electrical device indicated by the first operation state by using a mixed reality device; detecting an action of a first user by using the mixed reality device; in response to that an operation performed by the first user on the electrical device has been detected, performing the operation on an image model of the electrical device in the holographic image to obtain a second operation state after the electrical device is operated; and updating the first health state to a second health state indicated by the second operation state.

Abstract: A nucleic acid extraction element is provided, which is made of one or more of polyacrylate, polyamide, polyester, polyurethane, and polylactic acid (PLA), where the extraction element adsorbs nucleic acid in a first solution with a pH value of 2 to 6, and releases the nucleic acid in a second solution with a pH value of 8 to 12. Therefore, the extraction element can be used for extraction and detection of the nucleic acid, greatly shorten the time for extraction, cleaning, and purification of the nucleic acid, simplify operating steps for extraction and purification of the nucleic acid, and reduce consumption of equipment for extraction and purification of the nucleic acid.

Abstract: Modem deep neural network (DNN) models have many layers with a single layer potentially involving large matrix multiplications. Such heavy calculation brings challenges to deploy such DNN models on a single edge device, which has relatively limited computation resources. Therefore, multiple and even heterogeneous edge devices may be required for applications with stringent latency requirements. Disclosed in the present patent documents are embodiments of a model scheduling framework that schedules multiple models on a heterogeneous platform. Two different approaches, model first scheduling (MFS) and hardware first scheduling (HFS), are presented to allocate a group of models for a service into corresponding heterogeneous edge devices, including CPU, VPU and GPU. Experimental results prove the effectiveness of the MFS and HFS methods for improving the inference speed of single and multiple AI-based services.

Abstract: A semiconductor structure includes an alternating stack of insulating layers and composite layers, each of the composite layers includes a plurality of electrically conductive word line strips and a plurality of dielectric isolation structures, and each of the insulating layers has an areal overlap with each electrically conductive word line strip and each dielectric isolation structure within the composite layers within a memory array region in a plan view along a vertical direction, rows of memory openings arranged along the first horizontal direction, where each row of memory openings of the rows of memory openings vertically extends through each insulating layer within the alternating stack and one electrically conductive strip for each of the composite layers, and rows of memory opening fill structures located within the rows of memory openings, where each of the memory opening fill structures includes a vertical stack of memory elements and a vertical semiconductor channel.

Abstract: The present disclosure provides a parabolic tooth trace gear mechanism with a combined tooth profile of an end face circular arc and a parabola, which belongs to the field of gear transmission. The parabolic tooth trace gear mechanism with the combined tooth profile of the end face circular arc and the parabola includes a small wheel and a big wheel geared externally with parallel axes, tooth surfaces of the small wheel and the big wheel both have a parabolic tooth trace structure, and both are formed by the motion of a combined tooth profile curve of an end face along a contact point, contact lines are all parabolas after being unfolded along a cylindrical surface, and the parabola is determined by a meshing line parameter equation and basic design parameters such as a coincidence degree, the number of teeth, a transmission ratio and the like. When correctly installed, at least one pair of gear teeth of the small wheel and the big wheel, implement pure rolling meshing contact at a node.

Abstract: A light emitting device (LED) includes an n-doped semiconductor material layer, an active region including an optically active compound semiconductor layer stack configured to emit light located on the n-doped semiconductor material layer, a p-doped semiconductor material layer located on the active region, an anode contact contacting the p-doped semiconductor material layer, a reflector overlying and electrically connected to the anode contact, and a device-side bonding pad layer located on the reflector. The p-doped semiconductor material layer includes an electrically active region that is at least partially covered by the anode contact and an inactive region that an electrical conductivity less than 30% of the electrically active region.

Abstract: A system determines a timing matrix corresponding to inference times taken for a number of machine learning (ML) models to be executed by a number of processing resources of a computing device. The processing resources includes at least a first and a second type of processing resources. The system applies a service-specific model-first scheduling scheme or a service-specific hardware-first scheduling scheme to obtain corresponding service-specific mappings. The system determines a best mapping from the corresponding service-specific mappings. The system schedules each of the ML models to a corresponding processing resource from the processing resources according to the best mapping. The system executes the ML models using corresponding mapped processing resources.

Abstract: A method of generating an iso-surface mesh for a three-dimensional (3D) model is provided. In the method a plurality of voxels is generated in a bounding box of an input mesh, where a size of each of the plurality of voxels is based on a size of the bounding box of the input mesh and an iso-value. One or more iso-surfaces in each of the plurality of voxels are determined based on an unsigned distance function and the iso-value. One or more feature points in each of the plurality of voxels are determined based on the one or more iso-surfaces in the respective voxel and a constraint function. A feature mesh is generated based on the determined one or more feature points, and an edge flip operation is performed on each of a plurality of edges of the feature mesh to generate the iso-surface mesh.

Abstract: A spreader configured to be implemented in a die of a die casting system is disclosed. The spreader includes: a substrate; a hardening layer and an oxide layer. The hardening layer includes: a nitride and carbide layer disposed on the substrate and increasing hardness; and a white layer disposed on the nitride and carbide layer. The oxide layer includes: an inner transition layer disposed on the white layer and increasing resistance to oxidation; and an outer iron oxide layer disposed on the inner transition layer and increasing resistance to soldering during die casting of a part in the die casting system.

Abstract: A semiconductor memory structure includes a substrate having a device cell region and a contact forming region in proximity to the device cell region. A memory cell transistor is disposed within the device cell region. The memory cell transistor includes a gate and a charge storage structure between the gate and the substrate. The gate includes an extended portion within the contact forming region. A first spacer is disposed on a sidewall of the gate within the device cell region. A second spacer is disposed on a sidewall of the extended portion of the gate within the contact forming region. The second spacer is higher than the first spacer.