Weidner#
Index#
Let’s first import some packages:
[1]:
import os
import inspect
import shutil
import json
import torch
import pandas as pd
import pyaging as pya
Instantiate model class#
[2]:
def print_entire_class(cls):
source = inspect.getsource(cls)
print(source)
print_entire_class(pya.models.Weidner)
class Weidner(pyagingModel):
def __init__(self):
super().__init__()
def preprocess(self, x):
if self.reference_values is None:
return x
if isinstance(self.reference_values, torch.Tensor):
reference = self.reference_values.to(device=x.device, dtype=x.dtype)
else:
reference = torch.tensor(self.reference_values, device=x.device, dtype=x.dtype)
return torch.where(torch.isnan(x), reference, x)
def postprocess(self, x):
return x
[3]:
model = pya.models.Weidner()
Define clock metadata#
[4]:
model.metadata["clock_name"] = "weidner"
model.metadata["data_type"] = "DNA methylation" # Paper: DNA methylation changes at just three CpG sites
model.metadata["species"] = "Homo sapiens" # Paper: blood samples from healthy human donors
model.metadata["year"] = 2014
model.metadata["approved_by_author"] = "⌛"
model.metadata["citation"] = "Weidner, C. I., Lin, Q., Koch, C. M., et al. (2014). Aging of blood can be tracked by DNA methylation changes at just three CpG sites. Genome Biology, 15, R24."
model.metadata["doi"] = "https://doi.org/10.1186/gb-2014-15-2-r24"
model.metadata["notes"] = "Three-site whole-blood epigenetic-age estimator. The sites were selected from Illumina 27K blood profiles, and the final multivariate linear equation was fitted on targeted bisulfite-pyrosequencing beta values from 82 blood samples and validated in 69 independent samples."
model.metadata["research_only"] = None
model.metadata["tissue"] = ["whole blood"] # Paper: DNAm profiles derived from blood samples; model fitted on blood samples
model.metadata["predicts"] = ["biological age"] # Paper: signature provides a simple biomarker to estimate the state of aging in blood
model.metadata["training_target"] = ["chronological age"] # Paper: multivariate linear model to predict donor age
model.metadata["unit"] = ["years"] # Paper: Predicted age (in years)
model.metadata["model_type"] = "linear regression" # Paper: multivariate linear regression model
model.metadata["platform"] = ["Illumina 27K", "bisulfite sequencing"] # Paper: HumanMethylation27 BeadChip feature selection followed by pyrosequencing after bisulfite conversion
model.metadata["population"] = "adults" # Paper: 82 blood samples for training and independent validation set of 69 blood samples
model.metadata["journal"] = "Genome Biology"
model.metadata["last_author"] = "Wolfgang Wagner"
model.metadata["n_features"] = 3
model.metadata["citations"] = 973
model.metadata["citations_date"] = "2026-07-05"
Download clock dependencies#
[5]:
supplementary_url = "https://raw.githubusercontent.com/bio-learn/biolearn/180852e2bab473303cb85da627178b1695ee9d86/biolearn/data/Weidner.csv"
supplementary_file_name = "Weidner.csv"
os.system(f"curl -sL -o {supplementary_file_name} {supplementary_url}")
[5]:
0
Load features#
[6]:
df = pd.read_csv('Weidner.csv')
model.features = df['CpGmarker'].tolist()
Load weights into base model#
[7]:
weights = torch.tensor(df['CoefficientTraining'].tolist()).unsqueeze(0).float()
intercept = torch.tensor([38.0]).float()
[8]:
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#
[9]:
model.reference_values = None
Load preprocess and postprocess objects#
[10]:
model.preprocess_name = None
model.preprocess_dependencies = None
[11]:
model.postprocess_name = None
model.postprocess_dependencies = None
Check all clock parameters#
[12]:
pya.utils.print_model_details(model)
%==================================== Model Details ====================================%
Model Attributes:
training: True
metadata: {'approved_by_author': '⌛',
'citation': 'Weidner, Carola I., et al. "Aging of blood can be tracked by DNA '
'methylation changes at just three CpG sites." Genome biology '
'15.2 (2014): R24.',
'clock_name': 'weidner',
'data_type': 'methylation',
'doi': 'https://doi.org/10.1186/gb-2014-15-2-r24',
'notes': None,
'research_only': None,
'species': 'Homo sapiens',
'version': None,
'year': 2014}
reference_values: None
preprocess_name: None
preprocess_dependencies: None
postprocess_name: None
postprocess_dependencies: None
features: ['cg02228185', 'cg25809905', 'cg17861230']
base_model_features: None
%==================================== Model Details ====================================%
Model Structure:
base_model: LinearModel(
(linear): Linear(in_features=3, out_features=1, bias=True)
)
%==================================== Model Details ====================================%
Model Parameters and Weights:
base_model.linear.weight: tensor([[-26.4000, -23.7000, 164.7000]])
base_model.linear.bias: tensor([38.])
%==================================== Model Details ====================================%
Basic test#
[13]:
torch.manual_seed(42)
input = torch.randn(10, len(model.features), dtype=float)
model.eval()
model.to(float)
pred = model(input)
pred
[13]:
tensor([[ 70.9916],
[157.0004],
[155.2728],
[ 58.9885],
[ 38.9402],
[143.5184],
[396.4215],
[ 66.8971],
[130.9565],
[-26.1560]], dtype=torch.float64, grad_fn=<AddmmBackward0>)
Save torch model#
[14]:
torch.save(model, f"../weights/{model.metadata['clock_name']}.pt")
Clear directory#
[15]:
# 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: Weidner.csv