A Comprehensive Comparative Analysis of Deep Learning Based Feature Representations for Molecular Taste Prediction

被引:2
作者
Song, Yu [1 ,2 ,3 ]
Chang, Sihao [2 ,3 ]
Tian, Jing [2 ,3 ]
Pan, Weihua [2 ,3 ]
Feng, Lu [1 ]
Ji, Hongchao [2 ,3 ]
机构
[1] Zhengzhou Univ, Sch Agr Sci, State Key Lab Cotton Biol, Zhengzhou Res Base, Zhengzhou 450001, Peoples R China
[2] Guangdong Lab Lingnan Modern Agr, Shenzhen Branch, Shenzhen 518120, Peoples R China
[3] Chinese Acad Agr Sci, Agr Genom Inst Shenzhen, Genome Anal Lab, Minist Agr & Rural Affairs, Shenzhen 518120, Peoples R China
关键词
molecular feature representation; cheminformatics; taste prediction; machine learning; deep learning;
D O I
10.3390/foods12183386
中图分类号
TS2 [食品工业];
学科分类号
0832 ;
摘要
Taste determination in small molecules is critical in food chemistry but traditional experimental methods can be time-consuming. Consequently, computational techniques have emerged as valuable tools for this task. In this study, we explore taste prediction using various molecular feature representations and assess the performance of different machine learning algorithms on a dataset comprising 2601 molecules. The results reveal that GNN-based models outperform other approaches in taste prediction. Moreover, consensus models that combine diverse molecular representations demonstrate improved performance. Among these, the molecular fingerprints + GNN consensus model emerges as the top performer, highlighting the complementary strengths of GNNs and molecular fingerprints. These findings have significant implications for food chemistry research and related fields. By leveraging these computational approaches, taste prediction can be expedited, leading to advancements in understanding the relationship between molecular structure and taste perception in various food components and related compounds.
引用
收藏
页数:16
相关论文
共 45 条
[1]   A graph representation of molecular ensembles for polymer property prediction [J].
Aldeghi, Matteo ;
Coley, Connor W. .
CHEMICAL SCIENCE, 2022, 13 (35) :10486-10498
[2]   BitterSweet Forest: A Random Forest Based Binary Classifier to Predict Bitterness and Sweetness of Chemical Compounds [J].
Banerjee, Priyanka ;
Preissner, Robert .
FRONTIERS IN CHEMISTRY, 2018, 6
[3]   PredictSNP: Robust and Accurate Consensus Classifier for Prediction of Disease-Related Mutations [J].
Bendl, Jaroslav ;
Stourac, Jan ;
Salanda, Ondrej ;
Pavelka, Antonin ;
Wieben, Eric D. ;
Zendulka, Jaroslav ;
Brezovsky, Jan ;
Damborsky, Jiri .
PLOS COMPUTATIONAL BIOLOGY, 2014, 10 (01)
[4]   Prediction of bitterant and sweetener using structure-taste relationship models based on an artificial neural network [J].
Bo, Weichen ;
Qin, Dongya ;
Zheng, Xin ;
Wang, Yue ;
Ding, Botian ;
Li, Yinghong ;
Liang, Guizhao .
FOOD RESEARCH INTERNATIONAL, 2022, 153
[5]   FP-GNN: a versatile deep learning architecture for enhanced molecular property prediction [J].
Cai, Hanxuan ;
Zhang, Huimin ;
Zhao, Duancheng ;
Wu, Jingxing ;
Wang, Ling .
BRIEFINGS IN BIOINFORMATICS, 2022, 23 (06)
[6]   The receptors and cells for mammalian taste [J].
Chandrashekar, Jayaram ;
Hoon, Mark A. ;
Ryba, Nicholas J. P. ;
Zuker, Charles S. .
NATURE, 2006, 444 (7117) :288-294
[7]   Algebraic graph-assisted bidirectional transformers for molecular property prediction [J].
Chen, Dong ;
Gao, Kaifu ;
Duc Duy Nguyen ;
Chen, Xin ;
Jiang, Yi ;
Wei, Guo-Wei ;
Pan, Feng .
NATURE COMMUNICATIONS, 2021, 12 (01)
[8]  
Cho K., 2014, P 2014 C EMP METH NA, DOI 10.3115/v1/d14-1179
[9]   Gene-gene interaction detection with deep learning [J].
Cui, Tianyu ;
El Mekkaoui, Khaoula ;
Reinvall, Jaakko ;
Havulinna, Aki S. ;
Marttinen, Pekka ;
Kaski, Samuel .
COMMUNICATIONS BIOLOGY, 2022, 5 (01)
[10]   Bitter or not? BitterPredict, a tool for predicting taste from chemical structure [J].
Dagan-Wiener, Ayana ;
Nissim, Ido ;
Abu, Natalie Ben ;
Borgonovo, Gigliola ;
Bassoli, Angela ;
Niv, Masha Y. .
SCIENTIFIC REPORTS, 2017, 7