-
1.发明公开
公开(公告)号：EP4170662A1
公开(公告)日：2023-04-26
申请号：EP21879216.6
申请日：2021-09-22
发明人： XU, Tingyang , ZHANG, Jiying , YE, Fei , BIAN, Yatao , RONG, Yu
IPC分类号： G16C20/50
摘要： A method for determining interaction information, an interaction information prediction model training method, a device, and a medium are provided. The interaction information determining method includes: obtaining (201) basic information of a first target object, basic information of a second target object, and a target interaction information prediction model, the target interaction information prediction model being trained by using a global-level loss function and a key local-level loss function; and invoking (202) the target interaction information prediction model to process the basic information of the first target object and the basic information of the second target object, and obtaining target interaction information between the first target object and the second target object. Based on the foregoing process, the process of training the interaction information prediction model not only pays attention to the global information, but also pays attention to the key local information, so that the effect of training the interaction information prediction model is relatively good, and the accuracy of the determined interaction information between the first target object and the second target object is relatively high.
-
2.发明公开
公开(公告)号：EP3992976A1
公开(公告)日：2022-05-04
申请号：EP20877236.8
申请日：2020-09-24
发明人： RONG, Yu , HUANG, Wenbing , XU, Tingyang
IPC分类号： G16C20/30
摘要： Disclosed are a compound property prediction method and apparatus, and a computer device and a readable storage medium. Specifically, the method comprises: obtaining chemical structure information of a target compound, the chemical structure information comprising atoms and chemical bonds; generating a chemical structure graph corresponding to the chemical structure information according to the chemical structure information, the chemical structure graph comprising nodes corresponding to the atoms and edges corresponding to the chemical bonds; constructing original node features of the nodes and original edge features of the edges; performing a plurality of rounds of information propagation on the edges according to the original node features of the nodes and the original edge features of the edges to obtain propagation state information of the edges after the plurality of rounds of information propagation; obtaining target features of the edges according to the propagation state information; and predicting properties of the target compound according to the target features of the edges, and outputting a property prediction result of the target compound.
-
公开(公告)号：EP4393924A1
公开(公告)日：2024-07-03
申请号：EP22933129.3
申请日：2022-11-29
发明人： XUE, Ding , XU, Tingyang , HUANG, Junhong , LIU, Wei
IPC分类号： C07D487/04 , A61K31/52 , A61P35/00
CPC分类号： C07D487/04 , A61K31/4985 , A61K31/52 , A61P35/00 , C07D519/00
摘要： This application provides a compound. The compound is a compound as shown in Formula (I) or a stereoisomer, tautomer, deuterated compound, oxynitride, solvate, metabolite, pharmaceutically acceptable salt or prodrug of the compound having a structure as shown in Formula (I).
-
4.发明公开
公开(公告)号：EP4303878A1
公开(公告)日：2024-01-10
申请号：EP22882454.6
申请日：2022-08-31
发明人： BIAN, Yatao , XU, Tingyang
摘要： The present disclosure provides a method for training a molecular binding model, and a molecular screening method and apparatus, which can be applied in the field of intelligent medicine, and used to solve the problem of low molecular screening accuracy. The screening method comprises: using a molecular binding model to be trained, determining binding activity feature information, embedded feature information and eutectic feature information between a sample protein molecule and a sample candidate molecule, on the basis of protein feature information and molecular feature information; on the basis of the binding activity feature information, the embedded feature information and the eutectic feature information, determining a training loss of the molecule binding model to be trained; if the training loss meets a training target, outputting the molecular binding model as a trained molecular binding model, the trained molecular binding model being used to determine binding activity feature information between a target protein molecule and a target candidate molecule, so as to predict binding activity of a compound after the target protein molecule and the target candidate molecule are virtually bound.
-
5.发明公开
公开(公告)号：EP4266317A1
公开(公告)日：2023-10-25
申请号：EP22766169.1
申请日：2022-02-28
发明人： XU, Tingyang , HUANG, Junhong , XU, Shaoyong , TIAN, Li , CHEN, Xinde , LIU, Wei , HUANG, Junzhou , XUE, Ding , YU, Yang
摘要： Provided are a molecular structure reconstruction method and apparatus, a device, and a readable storage medium, relating to the field of machine learning. The method includes: 101: obtaining a set of structural data of a reference molecule, the reference molecule being a molecule that is active against a pre-set target, 102: performing structural separation on the obtained set of structural data to obtain a plurality of subsets of group data each corresponding to one of a plurality of molecular segment groups of the reference molecule, 103: performing feature processing on the group data of the molecular segment group to obtain a candidate segment for replacing the fragment segment, and 104: generating structural data of a reconstructed molecule based on the candidate segment and the side chain segment.
-
公开(公告)号：EP4201938A1
公开(公告)日：2023-06-28
申请号：EP22755561.2
申请日：2022-02-17
发明人： BAI, Qifeng , XU, Tingyang , FENG, Liya , LIU, Shuo , HUANG, Junzhou , LIU, Huanxiang , YAO, Xiaojun
IPC分类号： C07D495/04 , A61K31/4365 , A61P3/10
摘要： The present disclosure relates to the field of medical technologies, and provides a compound and use thereof as a glucagon receptor antagonist. The compound includes a compound A, or an isomer, metabolite, prodrug, pharmaceutically acceptable ester, or pharmaceutically acceptable salt of the compound A. The glucagon receptor antagonist in the present disclosure has a novel skeleton structure and a low half maximal inhibitory concentration on the glucagon receptor.
-
7.发明公开
公开(公告)号：EP4198991A1
公开(公告)日：2023-06-21
申请号：EP22766179.0
申请日：2022-02-28
发明人： XU, Tingyang , YU, Yang , RONG, Yu , LIU, Wei , HUANG, Junzhou , TU, Guiping , QIU, Yaping , CHENG, Xuemin
IPC分类号： G16B15/30
摘要： A method and an apparatus for processing molecular scaffold transitions, a computer-readable medium, an electronic device, and a computer program product are provided. The method for processing molecular scaffold transitions includes: generating, according to a connection graph structure corresponding to a reference drug molecule, an atomic latent vector corresponding to the reference drug molecule; performing atom masking processing on the atomic latent vector to obtain a scaffold latent vector and a sidechain latent vector included in the atomic latent vector; generating a target scaffold latent vector with a target transition degree between the scaffold latent vector and the target scaffold latent vector according to a spatial distribution of the scaffold latent vector; and generating a transitioned drug molecule according to the target scaffold latent vector and the sidechain latent vector.
-
8.发明公开
公开(公告)号：EP3992975A1
公开(公告)日：2022-05-04
申请号：EP20894594.9
申请日：2020-09-17
发明人： RONG, Yu , HUANG, Wenbing , XU, Tingyang
IPC分类号： G16C20/20
摘要： The present application relates to a compound property analysis method and apparatus, a compound property analysis model training method, and a storage medium, wherein same relate to the technical field of machine learning. The compound property analysis method comprises: acquiring a feature vector of a compound according to the molecular structure of the compound, wherein the feature vector comprises a node vector of each node and an edge vector of each edge; and processing the feature vector by means of a feature map extraction model branch, so as to obtain a map representation vector, and processing the map representation vector by means of a classification model branch, so as to obtain a property of the compound. By using the method, in the process of analyzing a property of a compound, a map representation vector which can precisely represent a feature of the compound is acquired on the basis of a map data structure of the compound, and a classification property of the compound is acquired on the basis of the map representation vector, thereby improving the precision of determining a classification property of a compound.
-
-
-
-
-
-
-
搜索结果
8 条记录 (耗时 0.026 秒)
