CVDWesterman

Index

1. Instantiate model class

2. Define clock metadata

3. Download clock dependencies

4. Load features

5. Load weights into base model

6. Load reference values

7. Load preprocess and postprocess objects

8. Check all clock parameters

9. Basic test

10. Save torch model

11. Clear directory

Let’s first import some packages:

import os
import inspect
import shutil
import json
import torch
import pandas as pd
import pyaging as pya

Instantiate model class

def print_entire_class(cls):
    source = inspect.getsource(cls)
    print(source)

print_entire_class(pya.models.CVDWesterman)

class CVDWesterman(LinearReferenceClock):
    def postprocess(self, x):
        """Logistic transform to a CVD risk probability."""
        return torch.sigmoid(x)

model = pya.models.CVDWesterman()

Define clock metadata

model.metadata["clock_name"] = "cvdwesterman"
model.metadata["data_type"] = "DNA methylation"  # Paper: DNA methylation cardiovascular risk score from blood cohorts.
model.metadata["species"] = "Homo sapiens"  # Paper: DNA methylation cardiovascular risk score from blood cohorts.
model.metadata["year"] = 2020
model.metadata["approved_by_author"] = "⌛"
model.metadata["citation"] = "Westerman, K. et al. Epigenomic assessment of cardiovascular disease risk and interactions with traditional risk metrics. Journal of the American Heart Association 9, e015299 (2020)."
model.metadata["doi"] = "https://doi.org/10.1161/jaha.119.015299"
model.metadata["notes"] = "Whole-blood DNA-methylation score for cardiovascular risk. The paper's final cross-study learner stacks cohort-specific elastic-net Cox models; the packaged pyaging implementation is a 235-CpG linear score followed by a sigmoid."
model.metadata["research_only"] = None
model.metadata["tissue"] = ["whole blood"]  # Paper: DNA methylation data ... collected using the HumanMethylation450 platform.
model.metadata["predicts"] = ["cardiovascular disease risk"]  # Paper: postprocess_name = 'sigmoid'
model.metadata["training_target"] = ["cardiovascular disease"]  # Paper: CVD events included coronary heart disease, stroke, and death from CVD.
model.metadata["unit"] = ["probability"]  # Paper: Logistic transform to a CVD risk probability.
model.metadata["model_type"] = "elastic net Cox ensemble"  # Paper: Penalized Cox proportional hazards regressions with the elastic net penalty.
model.metadata["platform"] = ["Illumina 450K"]  # Paper: Illumina HumanMethylation450 microarray platform
model.metadata["population"] = "adults"  # Paper: WHI, FHS-JHU, and LBC were the initial training cohorts.
model.metadata["journal"] = "Journal of the American Heart Association"
model.metadata["last_author"] = "José M. Ordovás"
model.metadata["n_features"] = 235
model.metadata["citations"] = 53
model.metadata["citations_date"] = "2026-07-05"

Download clock dependencies

os.system(f"curl -sL -o coefficients.csv https://raw.githubusercontent.com/bio-learn/biolearn/180852e2bab473303cb85da627178b1695ee9d86/biolearn/data/CVD_Westermann.csv")

0

Load features

df = pd.read_csv('coefficients.csv')
mask = df['CpGmarker'].astype(str).str.lower().isin(['intercept', '(intercept)'])
intercept_value = float(df.loc[mask, 'CoefficientTraining'].iloc[0]) if mask.any() else 0.0
coef_df = df.loc[~mask].reset_index(drop=True)
model.features = coef_df['CpGmarker'].tolist()

Load weights into base model

weights = torch.tensor(coef_df['CoefficientTraining'].tolist()).unsqueeze(0).float()
intercept = torch.tensor([intercept_value]).float()

base_model = pya.models.LinearModel(input_dim=len(model.features))

base_model.linear.weight.data = weights.float()
base_model.linear.bias.data = intercept.float()

model.base_model = base_model

Load reference values

model.reference_values = None

Load preprocess and postprocess objects

model.preprocess_name = None
model.preprocess_dependencies = None

model.postprocess_name = 'sigmoid'
model.postprocess_dependencies = None

Check all clock parameters

pya.utils.print_model_details(model)

%==================================== Model Details ====================================%
Model Attributes:

training: True
metadata: {'approved_by_author': '⌛',
 'citation': 'Westerman, Kenneth, et al. "Epigenomic assessment of '
             'cardiovascular disease risk and interactions with traditional '
             'risk metrics." Journal of the American Heart Association 9.8 '
             '(2020): e015299.',
 'clock_name': 'cvdwesterman',
 'data_type': 'methylation',
 'doi': 'https://doi.org/10.1161/JAHA.119.015299',
 'notes': None,
 'research_only': None,
 'species': 'Homo sapiens',
 'version': None,
 'year': 2020}
reference_values: None
preprocess_name: None
preprocess_dependencies: None
postprocess_name: 'sigmoid'
postprocess_dependencies: None
features: ['cg13077366', 'cg13679303', 'cg19214707', 'cg18593317', 'cg03463778', 'cg26626449', 'cg05630272', 'cg23522872', 'cg12588880', 'cg13074055', 'cg02655711', 'cg12624197', 'cg05843457', 'cg10963061', 'cg12121643', 'cg23613051', 'cg26504835', 'cg23063825', 'cg25188006', 'cg00887153', 'cg16574737', 'cg25283121', 'cg23239574', 'cg25564433', 'cg10612096', 'cg13903421', 'cg16268165', 'cg00600477', 'cg10516993', 'cg03930209']... [Total elements: 235]
base_model_features: None

%==================================== Model Details ====================================%
Model Structure:

base_model: LinearModel(
  (linear): Linear(in_features=235, out_features=1, bias=True)
)

%==================================== Model Details ====================================%
Model Parameters and Weights:

base_model.linear.weight: [0.0023171070497483015, 0.0011196619598194957, -0.013891435228288174, -0.002025559078902006, -0.009562727063894272, 0.06509598344564438, -0.020258523523807526, 0.0033463360741734505, -0.007966549135744572, -0.010243932716548443, -0.01568872109055519, 0.028809623792767525, 0.002753776963800192, 0.019885685294866562, 0.047365110367536545, 0.037816889584064484, 0.0031414860859513283, -0.004254682920873165, 0.002840817905962467, 0.002047969028353691, 0.0016441689804196358, 0.010059863328933716, 0.013813807629048824, 0.002602970926091075, 0.0023500178940594196, 0.004787160083651543, -0.006558055058121681, 0.01796160265803337, 0.016836676746606827, 0.0022038749884814024]... [Tensor of shape torch.Size([1, 235])]
base_model.linear.bias: tensor([0.])

%==================================== Model Details ====================================%

Basic test

torch.manual_seed(42)
input = torch.randn(10, len(model.features), dtype=float)
model.eval()
model.to(float)
pred = model(input)
pred

tensor([[0.3971],
        [0.6197],
        [0.7023],
        [0.5075],
        [0.5564],
        [0.5077],
        [0.5431],
        [0.4493],
        [0.6137],
        [0.3952]], dtype=torch.float64, grad_fn=<SigmoidBackward0>)

Save torch model

torch.save(model, f"../weights/{model.metadata['clock_name']}.pt")

Clear directory

# Function to remove a folder and all its contents
def remove_folder(path):
    try:
        shutil.rmtree(path)
        print(f"Deleted folder: {path}")
    except Exception as e:
        print(f"Error deleting folder {path}: {e}")

# Get a list of all files and folders in the current directory
all_items = os.listdir('.')

# Loop through the items
for item in all_items:
    # Check if it's a file and does not end with .ipynb
    if os.path.isfile(item) and not item.endswith('.ipynb'):
        os.remove(item)
        print(f"Deleted file: {item}")
    # Check if it's a folder
    elif os.path.isdir(item):
        remove_folder(item)

Deleted file: coefficients.csv