Paper Publications

Current position: Home > Scientific Research > Paper Publications

理工科高水平行业特色大学创新创业教育政策、模式与实践研究——以中国矿业大学为例

  • Release time:2024-07-20
  • Hits:

  • Impact Factor: 

    9.5

  • Journal: 

    Briefings in Bioinformatics

  • Key Words: 

    essential gene prediction; large language model; graph neural network; adversarial training; biological interpretation

  • Abstract: 

    Theidentificationandcharacterizationofessentialgenesarecentraltoourunderstandingofthecorebiologicalfunctionsineukaryoticorganisms,andhasimportantimplicationsforthetreatmentofdiseasescausedby,forexample,cancersandpathogens.Giventhemajorconstraintsintestingthefunctionsofgenesofmanyorganismsinthelaboratory,duetotheabsenceofinvitroculturesand/orgeneperturbationassaysformostmetazoanspecies,therehasbeenaneedtodevelopinsilicotoolsfortheaccuratepredictionorinferenceofessentialgenestounderpinsystemsbiologicalinvestigations.Majoradvancesinmachinelearningapproachesprovideunprecedentedopportunitiestoovercometheselimitationsandacceleratethediscoveryofessentialgenesonagenome-widescale.Here,wedevelopedandevaluatedalargelanguagemodel-andgraphneuralnetwork(LLM–GNN)-basedapproach,called‘Bingo’,topredictessentialprotein-codinggenesinthemetazoanmodelorganismsCaenorhabditiselegansandDrosophilamelanogasteraswellasinMusmusculusandHomosapiens(aHepG2cellline)byintegratingLLMandGNNswithadversarialtraining.Bingopredictsessentialgenesundertwo‘zero-shot’scenarioswithtransferlearning,showingpromisetocompensateforalackofhigh-qualitygenomicandproteomicdatafornon-modelorganisms.Inaddition,theattentionmechanismsandGNNExplainerwereemployedtomanifestthefunctionalsitesandstructuraldomainwithmostcontributiontoessentiality.Inconclusion,Bingoprovidestheprospectofbeingabletoaccuratelyinfertheessentialgenesoflittle-orunder-studiedorganismsofinterest,andprovidesabiologicalexplanationforgeneessentiality.

  • Indexed by: 

    Journal paper

  • Document Type: 

    J

  • Translation or Not: 

    no

  • Date of Publication: 

    2024-01-12

  • Included Journals: 

    SCI

Attachments:

Address:No1,Daxue Road,Xuzhou,Jiangsu,221116,P.R.China    Email: master@cumt.edu.cn
All Rights Reserved by China University of Mining and Technology