Abstract: The present disclosure provides a method, an apparatus, an electronic device, a storage medium and a program product for perceiving a road structure, and relates to the technical field of artificial intelligence and, in particular, to the technical field of deep learning and automatic driving. A specific implementation includes: determining, based on map data, a first perceiving result characterizing a road structure around a current position; determining, by a road structure perceiving model trained in advance, a second perceiving result characterizing the road structure around the current position; and generating a final perceiving result characterizing the road structure around the current position based on the first perceiving result and the second perceiving result.

Abstract: An mRNA drug, which reduces expression in the liver after in vivo delivery, and a preparation method thereof. The preparation method comprises: establishing double 3? UTR in a target mRNA, and introducing a miR-122 binding site between the double 3? UTR, so as to effectively reduce the expression of the mRNA drug, which is delivered by means of LNP, in the liver, thereby realizing a specific inhibitory effect of miR-122 while retaining an enhancement effect of miR-122 on mRNA stability and translation efficiency. The mRNA drug can effectively reduce the expression of mRNA in the liver, realizing the efficient expression of the mRNA drug in non-hepatocellular targeted cells or tissues and increasing the effective non-hepatocyte tropism of the mRNA drug.

Abstract: A cationic lipid compound, a preparation method and a use thereof, and a delivery system for mRNA are provided. The cationic lipid compound has a structure represented by formula (I): In formula (I): X is O or N; n is 2-4; m is 2-4; a is 0 or 1; R1 is a chain structure comprising a tertiary amine; R2 is a linear fatty acyl group or a branched chain fatty acyl group; R3 is a branched chain fatty acyl group. The cationic lipid compound is used for preparing a delivery system for mRNA.

Abstract: Provided a method for training a multi-task fusion detection model includes: obtaining a single-task detection model of each detection task in a detection task set, and obtaining an initial multi-task fusion detection model to be trained based on each single-task detection model; obtaining a training sampling set of the initial multi-task fusion detection model by obtaining a single-task sampling data set of each detection task, in which the training sample set includes a single-task sample and a multi-task sample; and training the initial multi-task fusion detection model according to the single-task sample and/or the multi-task sample until the training is completed, to obtain a trained target multi-task fusion detection model.

Abstract: A method and a system for calculating tumor neoantigen burden (TNB) are provided. The method includes: step S1: processing a normal sample and a tumor sample, sequencing a specific region and detecting somatic mutations in the sample; step S2: annotating and filtering the somatic mutations, and translating to obtain mutant peptide sequences of a patient; step S3: filtering the mutant peptide sequences to obtain neopeptide sequences based on a proteome of the normal sample; step S4, performing a human leukocyte antigen (HLA) typing analysis based on alignment data of the normal sample to obtain HLA genotypes of the sample; step S5, predicting binding affinities between the neopeptide sequences and the HLA genotypes to obtain specific neoantigens of the sample, and performing a weighted scoring on each of the specific neoantigens and calculating the TNB of the sample.

Abstract: A method and a platform for detecting an immunogenicity of a tumor neoantigen are provided. Specifically, the detection method includes the following steps: culturing human peripheral blood monocytes ex vivo for 13 days, adding an antigenic peptide fragment of human influenza virus and stimulating and activating cytokines, antigenic peptides, and immunoadjuvants during the 13 days, and finally conducting enzyme-linked immunospot (ELISPOT) chromogenic reaction and instrument-based scanning, counting, and analysis to detect the immunogenicity of tumor neoantigen. An application of the detection method and platform in biomedicine is provided. Compared with the prior art, the detection method and platform have advantages and characteristics of a short detection period, high convenience, low consumption of experimental cells, and low detection cost. Therefore, the detection method and platform can be used for ex vivo high-throughput assay for the immunogenicity of the tumor neoantigen.

Abstract: The present disclosure provides a method, an apparatus, an electronic device, a storage medium and a program product for perceiving a road structure, and relates to the technical field of artificial intelligence and, in particular, to the technical field of deep learning and automatic driving. A specific implementation includes: determining, based on map data, a first perceiving result characterizing a road structure around a current position; determining, by a road structure perceiving model trained in advance, a second perceiving result characterizing the road structure around the current position; and generating a final perceiving result characterizing the road structure around the current position based on the first perceiving result and the second perceiving result.

Abstract: A three-dimensional data augmentation method includes that an original two-dimensional image and truth value annotation data matching the original two-dimensional image are acquired, that the original two-dimensional image and two-dimensional truth value annotation data are transformed according to a target transformation element to obtain a transformed two-dimensional image and transformed two-dimensional truth value annotation data, that an original intrinsic matrix is transformed according to the target transformation element to obtain a transformed intrinsic matrix, that a two-dimensional projection is performed on three-dimensional truth value annotation data according to the transformed intrinsic matrix to obtain projected truth value annotation data, and that three-dimensional augmentation image data is generated according to the transformed two-dimensional image, the transformed two-dimensional truth value annotation data, and the projected truth value annotation data.

Abstract: A ranging method based on a laser radar system, a device and a readable storage medium are provided. The ranging method includes receiving ranging data obtained by measuring via a plurality of laser radars of a laser radar system, where the ranging data includes distances between the laser radars and obstacles in measurement ranges of the laser radars; establishing a three-dimensional coordinate model according to the ranging data; and determining a distance between a target unmanned device and each obstacle according to the three-dimensional coordinate model. The ranging method can realize a ranging calibration task for a larger distance range, thereby effectively improving a calibration efficiency and accuracy of the visual image algorithm, compared with the existing method of ranging calibration of the visual image algorithm by a single laser radar.

Abstract: An ADV includes a method to combine data from multiple sensors. The method compresses video data from a camera to generate compressed video data. The compressed video data are segmented. The method time synchronizes each segment of the compressed video data with data from other sensors. The method then combines each segment of the compressed video data with the corresponding time-synchronized sensor data for the other sensors. In one embodiment, each segment of the compressed video data is independently decodable. In another embodiment, each segment of the compressed video data includes a compressed video unit that is prepended with a buffered portion of the compressed video data that immediately precede the compressed video unit. The length of the compressed video unit is smaller than the length of the independently decodable segment to offer finer granularity in time synchronizing the compressed video data with the other sensor data with a tradeoff.

Abstract: A method and a system for calculating tumor neoantigen burden (TNB) are provided. The method includes: step S1: processing a normal sample and a tumor sample, sequencing a specific region and detecting somatic mutations in the sample; step S2: annotating and filtering the somatic mutations, and translating to obtain mutant peptide sequences of a patient; step S3: filtering the mutant peptide sequences to obtain neopeptide sequences based on a proteome of the normal sample; step S4, performing a human leukocyte antigen (HLA) typing analysis based on alignment data of the normal sample to obtain HLA genotypes of the sample; step S5, predicting binding affinities between the neopeptide sequences and the HLA genotypes to obtain specific neoantigens of the sample, and performing a weighted scoring on each of the specific neoantigens and calculating the TNB of the sample.

Abstract: A deep learning-based method for predicting a binding affinity between human leukocyte antigens (HLAs) and peptides includes: step S101: encoding HLA sequences; step S102: constructing a sequence of an HLA-peptide pair; step S103: constructing an encoding matrix of the HLA-peptide pair; step S104: constructing an affinity prediction model for HLA-peptide binding. The new method considers the effects of the protein sequences of HLAs and the sequences of the peptides on affinity strength and develops a deep learning-based method for predicting a binding affinity between HLAs and peptides.

Abstract: A method is disclosed for integrating multi-omics data to extract a microsatellite instability (MSI)-based neoantigen for immunotherapy. The method includes the following steps: S1, integrating DNA and RNA sequencing data of a patient to detect the microsatellite instability (MSI) of the patient accurately; S2, translating open reading frames (ORFs) influenced by the detected MSI to acquire an MSI proteome; S3, mapping the MSI proteome against a normal human proteome to acquire a sample-specific proteome; and S4, acquiring a sample neoantigen. The new method reduces the rate of false positives in MSI detection, which is especially relevant for improving the efficacy of current clinical immunotherapy.

Abstract: The present application provides autonomous vehicle based position detection method and apparatus, a device and a medium, where the method includes: identifying an obtained first visual perception image according to an underlying neural network layer in a slender convolution kernel neural network model to determine feature information of the target linear object image, and identifying the feature information of the target linear object image by using a high-level neural network layer in the slender convolution kernel neural network model to determine size information of the target linear object image; further, matching the size information of the target linear object image with preset coordinate system map information to determine a position of the autonomous vehicle. Embodiments of the present application can accurately determine the position of the autonomous vehicle.

Abstract: A plasmid vector for expressing mRNA in vitro, a construction method and an application thereof are provided. The plasmid vector includes a poly(adenyl deoxyribonucleotide) (poly(dA)) fragment formed by more than 30 adenyl deoxyribonucleotides at the 3?-end tail of a gene to be inserted for expression. The plasmid vector for expressing mRNA in vitro in the present invention can express a target protein gene and a poly(A) formed by 60 adenyl ribonucleotides, and the expressed mRNA directly possesses the poly(A) without additional tailing operation. In addition, the mRNA transcribed in vitro shows stronger mRNA stability and higher protein expression ability after being transfected into cells.

Abstract: A delivery system for mRNA nucleic acid drugs, a preparation method and an application thereof are provided. The delivery system includes lipid nanoparticles for loading one or more kinds of mRNA molecules, wherein the lipid nanoparticles are prepared from raw materials including an ionizable cationic lipid, a phospholipid auxiliary lipid, cholesterol, and a polyethylene glycol-derivatized phospholipid. In the mRNA nucleic acid drug targeted intracellular delivery system based on the non-viral carrier of the present invention, the mRNA is concentrated and loaded by the electrostatic interaction between the ionizable cationic lipid and the mRNA. Phospholipid auxiliary lipid component-mediated pH sensitivity and late endosomal escape enable mRNA nucleic acid drugs to be efficiently delivered to target cells and then released into the cytoplasm of the target cells for exerting a pharmacodynamic effect.

Abstract: The present disclosure provides a visual perception method, an apparatus, a device and a medium based on an autonomous vehicle, the method includes inputting an obtained first visual perception image collected by the autonomous vehicle into a first neural network model, recognizing multi-channel feature information of at least one target recognition object to be recognized, to eliminate redundant feature information in the first visual perception image; further, inputting the multi-channel feature information of the at least one target recognition object to be recognized into at least one sub-neural network model in a second neural network model respectively, to obtain at least one target recognition object; where there is a one to one correspondence between the target recognition object and the sub-neural network model.

Abstract: An ADV includes a method to combine data from multiple sensors. The method compresses video data from a camera to generate compressed video data. The compressed video data are segmented. The method time synchronizes each segment of the compressed video data with data from other sensors. The method then combines each segment of the compressed video data with the corresponding time-synchronized sensor data for the other sensors. In one embodiment, each segment of the compressed video data is independently decodable. In another embodiment, each segment of the compressed video data includes a compressed video unit that is prepended with a buffered portion of the compressed video data that immediately precede the compressed video unit.

Abstract: A method and system for extracting neoantigens for immunotherapy includes the following steps: step S1: acquiring conventional proteomes of tumor tissue and normal tissue samples; step S2: acquiring nucleotide polymer sequence libraries of the tumor tissue and normal tissue samples and a specific proteome of the tumor tissue sample; step S3: acquiring a plurality of candidate tumor-specific neoantigens based on the conventional and specific proteomes of the tumor tissue sample and molecular human leukocyte antigen (HLA) typing; and step S4: calculating the presence of the plurality of candidate tumor-specific neoantigens in the conventional proteomes and the nucleotide polymer sequence libraries of the tumor tissue and normal tissue samples, and acquiring tumor-specific neoantigens with a multiple of gene expression changes as a filter rule.

Abstract: A tumor neoantigen prediction platform and an application thereof in a neoantigen vaccine development system are provided. The present invention selects 55 HLA-A and HLA-B subtypes with a high proportion of Chinese population from a common database, then establishes a method for constructing cell lines expressing associated HLA subtypes, and subsequently analyzes the resulting HLA subtype cell line binding proteomes by protein mass spectrometry at an attomolar (10?18 molar) level; through very high-precision mass spectrometry of protein profiling presented on the surface of a single cell, the present invention builds a high-frequency HLA-binding polypeptide database for Chinese population; subsequently, a tumor neoantigen prediction algorithm is optimized by a prediction platform including the HLA-binding polypeptide database, thereby significantly improving the tumor neoantigen prediction accuracy.