基于异质图动态特征学习的药物重定位预测

打开文本图片集

Abstract：ObjectiveToaddress thechallenges faced by existing artificial intelligencemethodsinmodeling complex heterogeneousbiologicalntworks，particularlytheirlimitationsincapturingcollborativerelationshipsbetwnnodesandin extractinghigh-order topological semanticfeatures，weproposeanovel drugrepositioningpredictionmethodbasedon dynamicrepresentationlearningonheterogeneousgraphs.MethodsAheterogeneousbiological graphthatintegratesdrugs， diseases，andtheirinteractionrelationshipswasconstructed，basedonhichadynamicgatedattentionmodulewasdesigned toextractdiscriminativetopologicalfeaturesofdrugsanddiseasesbyincorporatingadynamicgraphatentionmechanism.A gatedresidual featurefusionmechanism wasdevelopedtopreciselyintegratestructuralandsemanticinformationfrom multiplesimilarity networks toreduce feature redundancyand information los，thereby enabling accurate predictionof drugdisease asociations. Results Experiments and case studies conducted on multiple drug datasets related to complex diseases demonstrated that theproposed method outperformed existing mainstream models in drug repositioning prediction. ConclusionThe proposedmethdcanefectivelymodelcomplexassociations inheterogeneous biologicalnetworks，ehance the acuracy of drug repositioning prediction，and provide importanttechnical support for precision treatment of complex diseases and development of medical artificial intelligence.

Keywords： complex biological networks; graph neural networks; gating mechanism; drug repositioning

在药物重定位预测中，如何高效地表征并挖掘原始数据中的信息和规律是提升建模精度的关键。（剩余18270字）