{"doi":"10.3390/molecules30234605","title":"Graph Neural Networks vs. Traditional QSAR: A Comprehensive Comparison for Multi-Label Molecular Odor Prediction","abstract":"<jats:p>Molecular odor prediction represents a fundamental challenge in computational chemistry with significant applications in fragrance design, food science, and chemical safety assessment. While traditional Quantitative Structure–Activity Relationship (QSAR) methods rely on hand-crafted molecular descriptors, recent advances in graph neural networks (GNNs) enable direct end-to-end learning from molecular graph structures. However, systematic comparison between these approaches for multi-label odor prediction remains limited. This study presents a comprehensive evaluation of traditional QSAR methods compared with modern GNN approaches for multi-label molecular odor prediction. Using the GoodScent dataset containing 3304 molecules with six high-frequency odor types (fruity, green, sweet, floral, woody, herbal), we systematically evaluate 23 model configurations across traditional machine learning algorithms (Random Forest, SVM, GBDT, MLP, XGBoost, LightGBM) with three feature-processing strategies and three GNN architectures (GCN, GAT, NNConv). The results demonstrate that GNN models achieve significantly superior performance, with GCN achieving the highest macro F1-score of 0.5193 compared to 0.4766 for the best traditional method (MLP with basic preprocessing), representing a 24.1% relative improvement. Critically, we discover that threshold optimization is essential for multi-label chemical classification. These findings establish GNNs as the preferred approach for molecular property prediction tasks and provide crucial insights for handling class imbalance in chemical informatics applications.</jats:p>","journal":"Molecules","year":2025,"id":638632,"datarank":0.29188652235829704,"base_score":1.9459101490553132,"endowment":1.9459101490553132,"self_citation_contribution":0.29188652235829704,"citation_network_contribution":0.0,"self_endowment_contribution":0.29188652235829704,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":6,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":null,"is_data_producer":false,"deposit_databanks":null,"is_oa":false,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":null,"fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":1658906,"name":"Xianfa Cai","orcid":null,"position":1,"is_corresponding":false},{"id":1586892,"name":"Jincheng Li","orcid":null,"position":2,"is_corresponding":false},{"id":1658905,"name":"Tengteng Wen","orcid":"0000-0002-9286-0336","position":0,"is_corresponding":false}],"reference_count":0,"raw_metadata":{"has_enrichment":true,"resolved":true,"title":"Graph Neural Networks vs. Traditional QSAR: A Comprehensive Comparison for Multi-Label Molecular Odor Prediction","abstract":"<jats:p>Molecular odor prediction represents a fundamental challenge in computational chemistry with significant applications in fragrance design, food science, and chemical safety assessment. While traditional Quantitative Structure–Activity Relationship (QSAR) methods rely on hand-crafted molecular descriptors, recent advances in graph neural networks (GNNs) enable direct end-to-end learning from molecular graph structures. However, systematic comparison between these approaches for multi-label odor prediction remains limited. This study presents a comprehensive evaluation of traditional QSAR methods compared with modern GNN approaches for multi-label molecular odor prediction. Using the GoodScent dataset containing 3304 molecules with six high-frequency odor types (fruity, green, sweet, floral, woody, herbal), we systematically evaluate 23 model configurations across traditional machine learning algorithms (Random Forest, SVM, GBDT, MLP, XGBoost, LightGBM) with three feature-processing strategies and three GNN architectures (GCN, GAT, NNConv). The results demonstrate that GNN models achieve significantly superior performance, with GCN achieving the highest macro F1-score of 0.5193 compared to 0.4766 for the best traditional method (MLP with basic preprocessing), representing a 24.1% relative improvement. Critically, we discover that threshold optimization is essential for multi-label chemical classification. These findings establish GNNs as the preferred approach for molecular property prediction tasks and provide crucial insights for handling class imbalance in chemical informatics applications.</jats:p>","is_dataset_classified":null,"base_score":1.791759469228055,"endowment":1.791759469228055,"datacite_reuse_total":0,"file_count":0,"downloads":0,"views":0,"has_version_chain":false,"is_dataset":false,"is_oa":false,"pmid":"41375203","pmcid":"PMC12693603","openalex_id":"https://openalex.org/W4416956641","authors":[],"funders":[{"funder_name":"Guangzhou Science, Technology and Innovation Commission","grant_id":"2024A04J5030","title":null}],"total_grants":1,"fwci":1.7093,"citation_percentile":0.84604556,"influential_citations":0,"citation_trend":[{"year":2026,"count":5}],"oa_status":"gold","license":"cc-by","oa_locations":[{"url":"https://www.mdpi.com/1420-3049/30/23/4605/pdf?version=1764646342","host_type":"journal"},{"url":"https://www.mdpi.com/1420-3049/30/23/4605/pdf?version=1764646342","host_type":"publisher"},{"url":"https://www.mdpi.com/1420-3049/30/23/4605/pdf","host_type":"publisher"},{"url":"https://doi.org/10.3390/molecules30234605","host_type":"journal"},{"url":"https://pubmed.ncbi.nlm.nih.gov/41375203","host_type":"repository"},{"url":"https://doaj.org/article/880bf0320431491e87ebb241f9ebb166","host_type":"repository"},{"url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC12693603/","host_type":"repository"},{"url":"https://europepmc.org/articles/PMC12693603","host_type":"Europe_PMC"},{"url":"https://europepmc.org/articles/PMC12693603?pdf=render","host_type":"Europe_PMC"}],"fields_of_study":["Advanced Chemical Sensor Technologies","Olfactory and Sensory Function Studies","Machine Learning in Bioinformatics","Quantitative Structure-Activity Relationship","Neural Networks, Computer","Odorants","Machine Learning","Algorithms","Graph Neural Networks"],"mesh_terms":["Machine Learning","Graph Neural Networks","Algorithms","Odorants","Neural Networks, Computer","Quantitative Structure-Activity Relationship"],"keywords":["Odor","Molecular graph","Graph","Artificial neural network","Quantitative structure–activity relationship","Property (philosophy)","Molecular descriptor","Multi-label Classification","Threshold Optimization","Graph Neural Network (Gnn)","Molecular Odor Prediction"],"sdg_mappings":[],"linked_datasets":[],"clinical_trials":[],"software_tools":[],"database_accessions":[],"source":"live","citation_network_status":"fetched"},"created_at":"2026-08-06T21:02:32.347975Z","pmid":null,"pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}