Due to outstanding performance in cheminformatics,machine learning algorithms have been increas-ingly used to mine molecular properties and biomedical big data.The performance of machine learning models is known to critically depend on the selection of the hyper-parameter configuration.However,many studies either explored the optimal hyper-parameters per the grid searching method or employed arbitrarily selected hyper-parameters,which can easily lead to achieving a suboptimal hyper-parameter configuration.In this study,Hyperopt library embedding with the Bayesian optimization is employed to find optimal hyper-parameters for different machine learning algorithms.Six drug discovery datasets,including solubility,probe-likeness,hERG,Chagas disease,tuberculosis,and malaria,are used to compare different machine learning algorithms with ECFP6 fingerprints.This contribution aims to evaluate whether the Bernoulli Naive Bayes,logistic linear regression,AdaBoost decision tree,random forest,sup-port vector machine,and deep neural networks algorithms with optimized hyper-parameters can offer any improvement in testing as compared with the referenced models assessed by an array of metrics including AUC,F1-score,Cohen's kappa,Matthews correlation coefficient,recall,precision,and accuracy.Based on the rank normalized score approach,the Hyperopt models achieve better or comparable perfor-mance on 33 out 36 models for different drug discovery datasets,showing significant improvement achieved by employing the Hyperopt library.The open-source code of all the 6 machine learning frame-works employed in the Hyperopt python package is provided to make this approach accessible to more scientists,who are not familiar with writing code.

Jun Zhang;Qin Wang;Weifeng Shen

School of Chemistry and Chemical Engineering,Chongqing University,Chongqing 401331,China;School of Chemistry and Chemical Engineering,Chongqing University of Science&Technology,Chongqing 401331,China;Chongqing Key Laboratory of Theoretical and Computational Chemistry,Chongqing 400044,China

中国化学工程学报（英文版）

[1]Jun Zhang;Qin Wang;Weifeng Shen-.Hyper-parameter optimization of multiple machine learning algorithms for molecular property prediction using hyperopt library)[J].中国化学工程学报（英文版）,2022(12):115-125

hyperopt,cheminformatics,Hyperopt,ECFP6

optimization,multiple,machine,learning,algorithms,molecular,property,prediction,using,library,Due,outstanding,performance,have,been,increas,ingly,used,mine,properties,biomedical,big,models,known,critically,depend,selection,configuration,However,many,studies,either,explored,parameters,grid,searching,method,employed,arbitrarily,selected,which,easily,lead,achieving,suboptimal,In,this,study,embedding,Bayesian,find,different,Six,drug,discovery,datasets,including,solubility,probe,likeness,hERG,Chagas,disease,tuberculosis,malaria,fingerprints,This,contribution,aims,evaluate,whether,Bernoulli,Naive,logistic,linear,regression,AdaBoost,decision,tree,random,forest,sup,port,vector,deep,neural,networks,optimized,offer,improvement,testing,compared,referenced,assessed,by,array,metrics,score,Cohen,kappa,Matthews,correlation,coefficient,recall,precision,accuracy,Based,rank,normalized,approach,better,comparable,showing,significant,achieved,employing,open,source,code,frame,python,package,provided,make,accessible,more,scientists,who,not,familiar,writing

0.553572