1 激发态释能和诊疗机理
2 荧光分子设计模型简介
2.1 分子特征描述符
2.2 基于BO算法的荧光分子设计
2.3 基于SVM算法的荧光分子设计
2.4 基于树模型的荧光分子设计
2.5 基于GA算法的荧光分子设计
2.6 基于蒙特卡洛方法的荧光分子设计
2.7 基于DNN算法的荧光分子设计
3 基于多种构效关系的荧光成像及诊疗分子设计
3.1 基于膜通透性的荧光成像分子设计
3.2 基于聚集诱导发光机理(AIE)的荧光成像分子设计
3.3 基于激发态能隙与pH的近红外荧光成像分子设计
3.4 基于ISC过程的荧光诊疗分子设计
3.5 基于其他性质的荧光诊疗分子设计
4 总结与展望
表1 基于ML与多种构效关系的荧光诊疗分子设计Table 1 Molecular design of fluorescent theranostic molecules based on ML with QSAR |
| Structure-function relationship | Material category | Theranostic Attribute | Algorithm | Advantage |
|---|---|---|---|---|
| Membrane perme- ability descriptor | BODIPY, etc. | Bioimaging | D-MPNN, DNN, B-PvsC, HTS, etc. | Increase of membrane permeability and biocompatibility |
| AIE/ACQ | Molecules with extended conjugation | Bioimaging, theranostic of multiple diseases | SVM, DNN, TD-DFT, etc. | Acceleration of the molecule discovery cycle, proposal of new mechanisms |
| ΔEHL | Near-infrared fluore- scent molecules | Deep-tissue bioimaging | GCNN, ETR, XGBoost, etc. | Increase of molecular backbone stability |
| pH | Oxazene, etc. | Deep-tissue bioimaging | ATTRNN, TD-DFT, etc. | PH-regulated fluorescence switching, disease-specific imaging |
| ISC | PS, TADF, RTP, etc. | PDT, Bioimaging | BO, VAE, HIPNN, etc. | Prolongation of material triplet-state lifetime, acceleration of materials discovery, Enhancement of precision theranostic performance |
| PTT | PTA, etc. | PTT, PAI | RF, GAN, GA, etc. | Reduction of the photothermal agent deve- lopment cycle, increase of photothermal efficiency |