LMF-CP: An Interpretable Multimodal Late-Fusion Framework for Compound Carcinogenicity Prediction
International Journal of Molecular Sciences · 21.08.2026 · 27 (16) · 7499
DOI: 10.3390/ijms27167499 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Molecular Sciences, online erschienen am 21.08.2026. Verfasst von Yingjie Zhu und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Mathematics and Statistics, Changchun University, Changchun 130022, China. Hauptthemen laut MEDLINE: Carcinogens, Deep Learning. Die Arbeit nennt 78 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Department of Education of Jilin Province, National Natural Science Foundation of China, Jilin Provincial Department of Education Research Project.
Abstract (Originalsprache)
Accurately predicting the carcinogenicity of compounds is of great significance for drug discovery, clinical drug safety, and chemical risk assessment. Traditional methods for assessing carcinogenicity rely on animal testing, which suffers from limitations such as time-consuming processes, high costs, significant interspecies differences, and low predictive throughput. In recent years, computational modeling-based prediction methods (such as Quantitative Structure–Activity Relationships, QSAR) have made some progress, but they still face challenges such as insufficient molecular feature information and poor model interpretability. To overcome these barriers, the multimodal deep learning framework LMF-CP (Late Multimodal Fusion of Carcinogenicity Prediction) is proposed to enhance the performance and interpretability of compound carcinogenicity prediction. First, to comprehensively characterize the structural and physicochemical properties of compounds, a multimodal representation system based on four molecular modalities is constructed, namely SMILES sequences, molecular fingerprints, molecular images, and molecular graph structures. Specifically, Text Convolutional Neural Network (TextCNN), Multi-Layer Perceptron (MLP), Visual Geometry Group Network (VGGNet), as well as Molecular Graph Attention Network (MGAT) are employed to process this information, respectively. Second, to integrate information from different molecular representations, a late-stage fusion strategy based on Lasso stacking is employed. On the test set, LMF-CP achieves an area under curve (AUC) of 0.828, an accuracy (ACC) of 0.782, an F1 score of 0.786, a sensitivity (SEN) of 0.786, and a specificity (SPE) of 0.779. In addition, this paper combines Shapley Additive Explanations (SHAP) analysis with Bemis–Murcko scaffold analysis to interpret the model results from two perspectives. Finally, a visual online platform for predicting the carcinogenicity of compounds is designed, providing a convenient tool for the rapid assessment of compound carcinogenicity and structural interpretation.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- School of Mathematics and Statistics, Changchun University, Changchun 130022, China.
ORCID-Profile: Liujie He, Xinjie Liang
Förderung
- Department of Education of Jilin Province
- National Natural Science Foundation of China
- Jilin Provincial Department of Education Research Project
Zitieren
APA
Zhu, Y., He, L., & Liang, X. (2026). LMF-CP: An Interpretable Multimodal Late-Fusion Framework for Compound Carcinogenicity Prediction. International Journal of Molecular Sciences, 27(16), 7499. https://doi.org/10.3390/ijms27167499
Vancouver
Zhu Y, He L, Liang X. LMF-CP: An Interpretable Multimodal Late-Fusion Framework for Compound Carcinogenicity Prediction. International Journal of Molecular Sciences. 2026;27(16):7499. doi:10.3390/ijms27167499
BibTeX
@article{Zhu2026,
author = {Zhu, Yingjie and He, Liujie and Liang, Xinjie},
title = {LMF-CP: An Interpretable Multimodal Late-Fusion Framework for Compound Carcinogenicity Prediction},
journal = {International Journal of Molecular Sciences},
year = {2026},
volume = {27},
number = {16},
pages = {7499},
doi = {10.3390/ijms27167499}
}
EUVIMED