Research fellow (Omics analysis, AI analysis for aging research)

About this position

[Background of the recruitment and description of the project]

• Background of the recruitment and description of the project

Aging does not progress uniformly throughout the body. The brain, immune system, cardiovascular system, liver, and other organs age at different rates. Because circulating proteins reflect the physiological states of multiple organs and cell types, combining plasma proteomics with artificial intelligence may enable the minimally invasive assessment of organ-specific biological age and functional decline.

This project aims to develop organ-age and organ-age-gap prediction algorithms applicable to Asian populations by integrating large-scale plasma proteomic datasets generated using the Olink and SomaScan platforms. The organ age gap is defined as the difference between an organ’s AI-predicted biological age and an individual’s chronological age. It provides a quantitative measure of whether a particular organ is aging faster or slower than expected.

Existing organ-age models have largely been developed using Western cohorts and specific proteomic platforms. Their reproducibility in Asian populations and their transferability between Olink and SomaScan therefore remain insufficiently established.

We will apply regularized regression, gradient boosting, domain calibration, and explainable AI approaches such as SHAP to construct models that are robust to population and measurement-platform differences. Proteins contributing to organ-age predictions will subsequently be linked to aging-related pathways, including inflammation, mitochondrial dysfunction, cellular senescence, chromatin regulation, and extracellular-matrix remodeling.

The long-term objective is to establish a research foundation for visualizing biological aging, evaluating health-related risks, identifying potential intervention targets, and ultimately developing molecular digital twins that represent individual biological states.

[Work content and job description]

The successful candidate will work closely with the principal investigator and collaborating researchers and will contribute to the following activities:

Data organization, quality control, and preprocessing of plasma proteomic datasets, primarily generated using Olink and SomaScan

Harmonization of protein identifiers, measurement units, missing values, demographic variables, and sample-collection metadata

Cross-platform protein mapping and calibration between Olink and SomaScan

Annotation of organ-enriched proteins using GTEx and other public biological resources

Development of organ-age prediction models using LASSO, Elastic Net, gradient boosting, and related machine-learning approaches

Evaluation of model generalizability using train/test splits, leave-one-cohort-out validation, and leave-one-platform-out validation

Assessment of predictive performance using MAE, R², intraclass correlation, rank correlation, and calibration metrics

Identification of organ-specific contributing proteins using SHAP values, model coefficients, and network-based analyses

Integration of predictive proteins with molecular signatures related to aging, inflammation, mitochondrial function, chromatin regulation, and extracellular-matrix remodeling

Biological interpretation using ICE mouse data, single-cell datasets, and publicly available transcriptomic resources

Development and documentation of reproducible analytical pipelines, metadata specifications, and evaluation protocols

Preparation of data visualizations, research reports, conference presentations, and scientific manuscripts

Participation in project meetings and interdisciplinary collaboration with life scientists and data scientists

This position offers an opportunity to work at the interface of aging biology, proteomics, and artificial intelligence. The successful candidate will address fundamental questions such as how organ aging can be quantified, how data from different proteomic technologies can be integrated, and how pro

• Assigned department

Existing departments

[Work location]

• Address

278-8510 Chiba 3rd Floor, Building 10, 2641 Yamazaki, Noda-shi,

[Number of hired]

Number of hired:2 person(s)