TABLE 2.
Key AI methodologies and advances across aging model organisms.
| Model system | AI technique | Dataset | Advancement | References |
|---|---|---|---|---|
| Yeast | Hierarchical clustering | Genomic | Unidentified longevity gene discovery | Chen et al. (2013) |
| CNN and CapsNet ensemble | Microfluidic time-lapse images | Automation of yeast classification | Ghafari et al. (2021) | |
| 2-layer predictive ML model with NN | Genomic | Effect of gene deletion, mitochondrial function and chromatin silencing on lifespan | Huang et al. (2012) | |
| ML predictor based on NET-FF | Colony-growth phenotypes | Uncovered uncharacterized proteins related to cellular aging | Rodríguez-López et al. (2023) | |
| CNN, LSTM, DeepLabV3+ | Microscopy image | Developed microfluidic platform for automated yeast lifespan classification | Aspert et al. (2022) | |
| Division detection model (CV), 18-layer ResNet, YOLOv3, linear regression respectively | Microfluidic time-lapse images | Developed microfluidic platform for automated yeast RLS measurement | Ghafari et al. (2022), Thayer et al. (2022), Xiao et al. (2024) | |
| C. elegans | SVR,InceptionResNetV2; EfficientNet-B0 model; InceptionV3; U-Net-based HydraNet & CNN-based WormNet respectively | Video, microscopic image | Lifespan prediction | Czaplewski et al. (2022), Lin et al. (2021), Martineau et al. (2020), Song et al. (2022), Yakimovich and Galimov (2021) |
| RF | Image and video | Revealed pharyngeal and intestinal deterioration as key predictors of mortality | Kern et al. (2024) | |
| RF | DrugAge | Uncovered lifespan-extending compounds | Barardo et al. (2017), Ribeiro et al. (2023) | |
| Mol2vec | ChEMBL and ZINC | Discovered potent mitophagy inducers with anti-aging potential | Xie et al. (2022) | |
| Supervised ML pipeline | Multi-omics | Revealed novel targets for neurodegenerative disease | Truter et al. (2022) | |
| Att-EfficientNet, Faster R-CNN; Bimodal neural Network; ensemble of CNN + LSTM + GRU + transformer models, ResNet18 + LSTM respectively |
Fluorescence microscopy image, Synthetic image | Lifespan assay automation-classification into lifespan stages; tracking movement; lifespan termination prediction; classification into dead or alive worm respectively | Bates et al. (2022), García Garví et al. (2021), García-Garví et al. (2023a), García-Garví et al. (2023b), Song et al. (2022) | |
| Drosophila melanogaster | Regression and supervised ML models, 1D CNN | snRNA-seq data | Aging clock development- biological age estimation, sex- based differences in aging | Lu et al. (2023), Tennant et al. (2024) |
| VZI-ACP + CNN | Numerical count of sleep pattern | Lifespan estimation | Zhang et al. (2021) | |
| UNet + CNN | Video | Automated analysis of cardiac dysfucntion and dynamics | Melkani et al. (2024) | |
| SCENIC; XGBoost | Gene expression profile; RNA-seq data respectively | Unveiled biology of brain aging- revealed an exponential RNA decline with age; discovered 50 conserved aging-related genes respectively | Davie et al. (2018), Webb et al. (2021) | |
| SVM | Gene expression profile | Discovered role of mitochondria in aging | Zhang et al. (2008) | |
| 4 tree-based ensemble ML models- BRF + EEC + XGB + CAT; RF respectively | 9 biological features + GO terms; DGRP respectively | Relationship between dietary restriction and aging- identified 7 genes as both age- and diet-related, identified key longevity-associated metabolites respectively | Hilsabeck et al. (2023), Vega Magdaleno et al. (2022) | |
| Mice | NSC; XGBoost | Physiological and biochemical measurements; nanoparticle movement data | Lifespan prediction- physiological age; chronological brain age | McKenna et al. (2021), Swindell et al. (2008) |
| GPR + Logistic Regression; PCA + Partial Least Squares Discriminant Analysis + Hierarchical Clustering Analysis + Multivariate Receiver Operating Characteristic analysis + SVM respectively |
MRI images; Mass spectrometry imaging (MSI) data respectively | Brain study- brain-age prediction; aging-induced metabolic changes in specific brain regions respectively | Brusini et al. (2022), Vallianatou et al. (2021) | |
| Elastic net regression; RF respectively | RRBS data; Frailty Indices respectively | Aging clock- DNA methylation pattern analysis to explore the relationship between aging and epigenetic modifications; biological age + lifespan prediction respectively | Levine et al. (2020), Schultz et al. (2020) | |
| LASSO regression + SVM-RFE | Metabolomic Data | Biomarker discovery- metabolic | Shi et al. (2021) | |
| RF | Transcriptomic data | Examined age-related gene expression changes | Palmer et al. (2021) | |
| SVM-C | Microscopic image | Examined age-related morphological changes in microglia | Choi et al. (2022) | |
| ATR-FTIR + LDA + SVM | Infrared spectroscopic data | Blood plasma exchange on liver tissues; ileum and colon respectively | Ceylani et al. (2023), Teker et al. (2023) | |
| Human | RF + XGB + DNN + SVR | Clinical biomarker | Comparative analysis between conventional statistical models with 4 AI models | Bae et al. (2021) |
| XGBoost; Elastic net; ResNet-based CNN + transfer learning respectively | 44 clinical and physiological features; cerebrospinal fluid (CSF) proteomics data; T1-weighted MRI scans respectively | Age prediction- biological age; chronological age; brain age respectively | Jonsson et al. (2019), Melendez et al. (2024), Wang et al. (2022) | |
| Guided Autoencoder; Inception-v3 respectively | Blood immune biomarkers; retinal fundus image respectively | Aging clock- predicted aging related chronic inflammation; chronological age prediction from retina respectively | Ahadi et al. (2023), Sayed et al. (2021) | |
| RFR; ElasticNet Regression + SVM + kNN + RF + Deep Feature Selection respectively |
Blood Samples; Muscle gene expression data respectively |
Circular RNAs as aging biomarkers; 20 aging-related genes and pathways for muscle aging respectively |
Mamoshina et al. (2018), Wang et al. (2020) | |
| XGBoost + SVM + Logistic Regression | Human protein features | Classified proteins into aging-related and non-aging-related categories | Kerepesi et al. (2018) | |
| linear regression models + SHapley Additive exPlanations + Bayesian networks | Gene expression data | Identified 57 novel longevity-associated genes and key metabolic pathways influencing lifespan | Kulaga et al. (2021) | |
| PandaOmics | Omics data | Discovered genes with both anti-aging and anti-cancer potential | Pun et al. (2023) | |
| LASSO + RF + SVM-RFE | Gene expression data | Identification of aging-related genes linked to age-induced cardiac failure | Yu et al. (2024) |
ML, machine learning; DL, deep learning; NN, neural network; NET-FF, feedforward neural network; CNN, convolutional neural network; LSTM, Long Short-Term Memory; CAPSNET, capsule network; CV, computer vision; YOLOV3, you only look once; Version 3; RF, random forest; SVR, support vector regression; ResNet, Residual Network; Faster R-CNN, Faster Region-based Convolutional Neural Network; ATT- Efficient NET, Attention-based EfficientNet; GRU, gated recurrent unit; SVM, support vector machine; SCENIC, Single-Cell Regulatory Network Inference and Clustering; XGBoost, Extreme Gradient Boosting; RNA-seq, RNA, sequencing; UNet, U-shaped Convolutional Neural Network; BRF, balanced random forest; DGRP, drosophila genetic reference panel; EEC, easy ensemble classifier; CAT, Categorical Boosting (CatBoost); NSC, nearest shrunken centroid; VZI-ACP, Zero-inflated autoregressive conditional Poisson; DNN, deep neural network; LASSO, least absolute shrinkage and selection operator; SVM-RFE, Support Vector Machine - Recursive Feature Elimination; ATR-FTIR, Attenuated Total Reflectance - Fourier Transform Infrared Spectroscopy; LDA, linear discriminant analysis; SVM-C, support vector machine with cost parameter; GPR, gaussian process regression; PCA, principal component analysis.