LinAge Scientific Foundation

LinAge is designed to calculate an individual’s biological age based on various biometric inputs. It provides insights into a person’s health status by comparing their biological age to their chronological age. The algorithm is based on peer-reviewed scientific research by Dr. Sheng Fong and colleagues. It has been validated by demonstrating consistent ability to predict all-cause mortality and stratify survival trajectories across diverse populations, showcasing predictive power comparable to or exceeding other established aging clocks.
For practical use cases, read more about our LinAge Calculator. For API documentation and technical details, refer to the LinAge Calculator API. Or read about our entire Biometric Calculator suite.
Key Publications
Fong S, Pabis K, Latumalea D, Dugersuren N, Unfried M, Tolwinski N, Kennedy B, Gruber J. “Principal component-based clinical aging clocks identify signatures of healthy aging and targets for clinical intervention.” Nat Aging. 2024 Aug; 4(8):1137-1152. doi: 10.1038/s43587-024-00646-8. Epub 2024 Jun 19. PMID: 38898237; PMCID: PMC11333290.
This study introduces a clinical aging clock (PCAge) built using principal component analysis on routine clinical data to distinguish healthy from unhealthy aging trajectories. It identifies signatures (such as metabolic dysregulation, cardiac and renal dysfunction, and inflammation) that not only predict poor aging outcomes but also correspond to pathways that can be modified by existing drugs. From PCAge, the authors derive LinAge, a streamlined clock with similar predictive accuracy but requiring far fewer measurements. They even retrain a version tailored to the CALERIE caloric-restriction study (CALinAge), showing that two years of mild caloric restriction significantly reduces biological age.
Fong S, Denisov KA, Nefedova AA, Kennedy BK, Gruber J. “LinAge2: providing actionable insights and benchmarking with epigenetic clocks. NPJ Aging.” 2025 Apr 23; 11(1):29. doi: 10.1038/s41514-025-00221-4. PMID: 40268972; PMCID: PMC12019333.
LinAge2 is a refined, clinically based “aging clock” developed to better predict mortality and functional aging than both chronological age and existing epigenetic clocks, like PhenoAge DNAm and DunedinPoAm. When tested using NHANES data, it outperformed chronological age and several other clocks in forecasting 10- and 20-year all-cause mortality, with higher predictive accuracy (AUC values). Lower LinAge2 biological age correlated with better healthspan indicators—like cognitive performance, gait speed, and the ability to perform everyday tasks. Crucially, LinAge2’s design—based on principal component analysis of clinical parameters—makes it more interpretable and potentially actionable, letting clinicians identify specific physiological contributors to accelerated aging and tailor interventions accordingly.
The first paper (2024) introduced PCAge, a clinical aging clock built from principal component analysis of standard lab measures, and from it derived LinAge, a simpler but accurate version that could even detect the age – slowing effects of caloric restriction. It showed that key physiological signatures of aging (like metabolic, cardiac, renal, and inflammatory markers) are predictive of poor outcomes but also potentially modifiable. Building on that foundation, the second paper (2025) developed LinAge2, a refined model trained on broader NHANES data to improve mortality prediction over both chronological age and earlier clocks, including LinAge. LinAge2 not only outperformed previous models in predicting 10- and 20-year survival but also offered interpretability, linking biological age shifts to specific physiological systems and actionable interventions.
Algorithm Overview
LinAge uses a principal component-based approach to calculate biological age from clinical biomarkers and health assessments. The algorithm:
- Standardizes biomarkers using sex-specific median values and median absolute deviations
- Calculates derived indices including comorbidity index (LinAgeCI), healthcare use index (LinAgeHUI), and self-health index (LinAgeSHI)
- Applies sex-specific weights to each standardized biomarker
- Computes age acceleration using weighted biomarker scores
- Calculates biological age by adding age acceleration to chronological age
The algorithm incorporates both objective clinical measurements and subjective health assessments to provide a comprehensive view of biological aging.
Required Biomarkers
While LinAge requires only chronological age and sex to initiate a calculation, for accurate and meaningful results, it is highly recommended to provide as many of the following biomarkers and health assessments as possible. The algorithm is designed to accept any combination of available parameters, with missing values contributing a median or zero z-score to the calculation, though this will reduce accuracy.
Core Demographics
- Age (chronological age in years) - Required
- Sex (male/female) - Required
Blood Biomarkers
- Albumin (g/dL)
- Alkaline Phosphatase (ALP) (U/L)
- Alanine Aminotransferase (ALT) (U/L)
- Aspartate Aminotransferase (AST) (U/L)
- Basophils (K/uL and %)
- Bilirubin (mg/dL)
- Blood Urea Nitrogen (BUN) (mg/dL)
- Body Mass Index (BMI) (kg/m²)
- Calcium (mg/dL)
- Chloride (mEq/L)
- Carbon Dioxide (CO2) (mEq/L)
- Cotinine (ng/mL)
- Creatinine (mg/dL)
- C-Reactive Protein (CRP) (mg/L)
- Diastolic Blood Pressure (DBP) (mmHg)
- Eosinophil Percentage (%)
- Eosinophils (K/uL) - LinAge2 only
- Iron (ug/dL)
- Ferritin (ug/L)
- Fibrinogen (g/L) - LinAge1 only
- Folate (B9) (ng/mL)
- Gamma-Glutamyl Transferase (GGT) (U/L) - LinAge1 only
- Globulin (g/dL)
- Glucose (mg/dL)
- Hemoglobin A1c (%)
- Hematocrit (%)
- Hemoglobin (g/dL)
- High-Density Lipoprotein (HDL) (mg/dL) - LinAge1 only
- Heart Rate (bpm)
- Potassium (mEq/L)
- Lactate Dehydrogenase (LDH) (U/L)
- Low-Density Lipoprotein (LDL) (mg/dL) - LinAge1 only
- Low-Density Lipoprotein (LDL) Calculated (mg/dL) - LinAge2 only
- Lymphocytes (K/uL and %)
- Mean Corpuscular Hemoglobin (MCH) (pg)
- Mean Corpuscular Hemoglobin Concentration (MCHC) (g/dL)
- Mean Corpuscular Volume (MCV) (fL)
- Mean Platelet Volume (MPV) (fL)
- Monocytes (K/uL and %)
- Sodium (mEq/L)
- Neutrophils (K/uL and %)
- N-terminal Pro-brain Natriuretic Peptide (NT-proBNP) (pg/mL)
- Phosphorus (mg/dL)
- Platelets (K/uL)
- Protein (g/dL)
- Red Blood Cell Count (M/uL)
- Red Blood Cell Distribution Width (RDW) (%)
- Systolic Blood Pressure (SBP) (mmHg)
- Total Cholesterol (mg/dL) - LinAge1 only
- Triglycerides (mg/dL) - LinAge1 only
- Total Iron Binding Capacity (TIBC) (umol/L)
- Transferrin Saturation (%)
- Uric Acid (mg/dL)
- Vitamin B12 (pg/mL)
- White Blood Cell Count (WBC) (K/uL)
Urine Biomarkers
- Albumin (Urine) (mg/dL) - LinAge2 only
- Creatinine (Urine) (mg/dL) - LinAge2 only
- Albumin-Creatinine Ratio (Urine) (mg/g) - Calculated automatically if both urine values provided
Health Assessments
- General Health (“good” or better, “fair”, or “poor”)
- Current Health (“better” than, about the “same” as, or “worse” than last year)
- Healthcare Use Index (number of doctor visits in past 12 months, excluding overnight hospitalizations)
Comorbidity Diagnoses
- Anemia (AN_DX)
- Angina (AP_DX)
- Arthritis (AR_DX)
- Asthma (AS_DX)
- Malignancy (CA_DX)
- Chronic Bronchitis (CB_DX)
- Coronary Heart Disease (CHD_DX)
- Cognitive Impairment (CI_DX)
- Diabetes Mellitus (DM_DX)
- Emphysema (EMPH_DX)
- Hypertension (HTN_DX)
- Liver Disease (LD_DX)
- Obesity (OB_DX)
- Osteoporosis (OP_DX)
- Previous Hip Fracture (PHF_DX)
- Previous Myocardial Infarction (PMI_DX)
- Previous Overnight Hospitalization (POH_DX)
- Previous Stroke (PS_DX)
- Previous Spine Fracture (PSF_DX)
- Previous Wrist Fracture (PWF_DX)
- Renal Impairment (RI_DX)
- Thyroid Disease (TD_DX)
Derived Indices
LinAge calculates three derived indices that enhance the biological age assessment:
LinAge Comorbidity Index (LinAgeCI)
Quantifies the total burden of 22 specific chronic diseases by summing the number of reported diagnoses and dividing by 22. Can be provided directly or calculated automatically from individual diagnostic parameters.
LinAge Healthcare Use Index (LinAgeHUI)
Records the number of times an individual has seen a healthcare professional in the past 12 months, excluding overnight hospitalizations.
LinAge Self-Health Index (LinAgeSHI)
A weighted score derived from subjective health assessments, incorporating both general health status and perceived health changes over the past year.
Validation and Limitations
LinAge has been validated in multiple studies:
- NHANES cohort: Validated across diverse U.S. populations
- CALERIE study: Demonstrated effectiveness in caloric restriction interventions
- Mortality prediction: Strong association with all-cause mortality
- Clinical outcomes: Predicts health trajectories and intervention effectiveness
Current Limitations:
- Requires comprehensive clinical data for optimal accuracy
- May be influenced by acute illness or inflammation
- Limited validation in very young (<20) or very old (>85) populations
- Results should be interpreted in clinical context
The algorithm performs best with comprehensive biomarker data but can provide estimates with minimal parameters.
Technical Implementation
Our LinAge implementation includes:
- Sex-specific calculations: Separate median values, deviations, and weights for males and females
- Unit conversions: Automatic conversion between different measurement units
- Derived calculations: Automatic computation of comorbidity and health indices
- Validation checks: Negative biological ages are rejected
- Flexible input: Accepts any combination of available parameters
The algorithm returns both LinAge1 and LinAge2 variants, with the LinAge2 being the recommended final result.
LinAge1 (Original)
- Based on 2024 publication
- Initial clinical aging clock implementation
LinAge2 (Recommended)
- Based on 2025 publication with improved mortality prediction
- Enhanced interpretability and clinical utility
- Used as the primary biological age result
API Outputs:
result.biological
: Final biological age (uses LinAge2)result.LinAge1
: Raw LinAge1 algorithm resultresult.LinAge2
: Raw LinAge2 algorithm result (recommended)
Usage Guidelines
When to use LinAge:
- Adults aged 20-85 years
- Comprehensive clinical laboratory results available
- Assessment of biological aging and health trajectories
- Research studies of aging interventions
- Preventive medicine and personalized health planning
Interpretation:
- LinAge < Chronological Age: Younger biological age
- LinAge > Chronological Age: Older biological age
- Difference: Years of accelerated or decelerated aging
Clinical Considerations:
- Results should be interpreted by healthcare professionals
- Consider alongside other clinical assessments
- Not a diagnostic tool for specific diseases
- May vary with acute illness, inflammation, or medication use
- Comorbidity index provides additional clinical context
- Self-health assessments enhance subjective health evaluation
Minimum Requirements:
- Age and sex are required for basic calculation
- More comprehensive results require additional biomarkers
- Derived indices enhance accuracy when available