Abstract
Recent advances in lipidomics and machine learning have been harnessed to explore biological age prediction in individuals. This study delves into age acceleration patterns, entropy, and the role of dolichol as a potential aging biomarker. We present a novel aging clock in combination with explainable AI that utilizes lipid composition of the prefrontal cortex to predict biological age of individuals without known neurological conditions, as well as in those with autism, schizophrenia, or Down syndrome. Significant age acceleration was observed in individuals with autism, with a higher acceleration after the age of 40. In addition, entropy increases significantly around the age of 40, indicative of mevalonate pathway dysregulation. These findings underscore the feasibility of predicting biological age using lipidomics data, opening avenues for feature research into the intricate relationship between lipid alterations and aging of the prefrontal cortex, while providing valuable insights into the associated molecular mechanisms.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Abbreviations
- SHAP
- SHapley Additive exPlanations
- ND
- without neurological disorder
- DS
- Down syndrome
- ASD
- Autism Spectrum Disorder
- SZ
- schizophrenia
- PMI
- post-mortem interval
- LM IDs
- LIPID MAP IDs
- m/z
- molecular weight
- RT
- retention time
- FA
- fatty acyls
- GL
- glycerolipids
- GP
- glycerophospholipids
- SP
- sphingolipids
- ST
- sterol lipids
- PK
- polyketides
- PCA
- principal component analysis
- SVD
- singular value decomposition
- MAD
- mean absolute deviation
- ME
- mean error
- r
- Pearson correlation coefficient
- RMSE
- root mean squared error
- SE
- standard error