ReedBMI#

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:

[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.ReedBMI)

class ReedBMI(LinearReferenceClock):
    pass


[3]:

model = pya.models.ReedBMI()

Define clock metadata#

[4]:

model.metadata["clock_name"] = "reedbmi"
model.metadata["data_type"] = "DNA methylation"  # Paper: DNA methylation scores for BMI
model.metadata["species"] = "Homo sapiens"  # Paper: ARIES cohort of mothers and children
model.metadata["year"] = 2020
model.metadata["approved_by_author"] = "⌛"
model.metadata["citation"] = "Reed, Zoe E., et al. \"The association of DNA methylation with body mass index: distinguishing between predictors and biomarkers.\" Clinical Epigenetics 12 (2020): 50."
model.metadata["doi"] = "https://doi.org/10.1186/s13148-020-00841-5"
model.metadata["notes"] = "Weighted blood-DNA-methylation score built from published BMI EWAS effect estimates and evaluated across the ARIES life course; it is a biomarker associated with concurrent BMI, not a calibrated prediction in kilograms."
model.metadata["research_only"] = None
model.metadata["tissue"] = ["whole blood", "cord blood"]  # Paper: DNA methylation measured in peripheral blood; at birth, childhood, adolescence, pregnancy and middle age
model.metadata["predicts"] = ["BMI methylation score"]  # Paper: methylation score for BMI
model.metadata["training_target"] = ["body mass index"]  # Paper: 135 CpG sites from ... meta-analysis of DNA methylation and BMI
model.metadata["unit"] = ["unitless"]  # Paper: CpG methylation levels were multiplied by published effect estimates and summed; analyses report effects per SD of the score.
model.metadata["model_type"] = "weighted methylation aggregation"  # Paper: multiplying ... methylation levels ... with corresponding published effect estimates and then summing
model.metadata["platform"] = ["Illumina 450K"]  # Paper: ARIES methylation profiles
model.metadata["population"] = "all ages"  # Paper: mothers and children ... at multiple time points
model.metadata["journal"] = "Clinical Epigenetics"
model.metadata["last_author"] = "Gibran Hemani"
model.metadata["n_features"] = 135
model.metadata["citations"] = 86
model.metadata["citations_date"] = "2026-07-05"

Download clock dependencies#

[5]:

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

[5]:

0

Load features#

[6]:

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#

[7]:

weights = torch.tensor(coef_df['CoefficientTraining'].tolist()).unsqueeze(0).float()
intercept = torch.tensor([intercept_value]).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': 'Reed, Zoe E., et al. "Perinatal and childhood epigenetic '
             'biomarkers of body composition." Clinical Epigenetics 12.1 '
             '(2020): 156.',
 'clock_name': 'reedbmi',
 'data_type': 'methylation',
 'doi': 'https://doi.org/10.1186/s13148-020-00841-5',
 'notes': None,
 'research_only': None,
 'species': 'Homo sapiens',
 'version': None,
 'year': 2020}
reference_values: None
preprocess_name: None
preprocess_dependencies: None
postprocess_name: None
postprocess_dependencies: None
features: ['cg00108715', 'cg00222799', 'cg00574958', 'cg01130991', 'cg01243823', 'cg01368219', 'cg01455178', 'cg01526748', 'cg01597398', 'cg01671681', 'cg01751802', 'cg01798813', 'cg01881899', 'cg02286155', 'cg02571142', 'cg02711608', 'cg02716826', 'cg03078551', 'cg03218374', 'cg03500056', 'cg03682690', 'cg03725309', 'cg04011474', 'cg04286697', 'cg04483863', 'cg04557677', 'cg04816311', 'cg04927537', 'cg05119988', 'cg06192883']... [Total elements: 135]
base_model_features: None

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

base_model: LinearModel(
  (linear): Linear(in_features=135, out_features=1, bias=True)
)

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

base_model.linear.weight: [0.007000000216066837, 0.007000000216066837, -0.019999999552965164, 0.009999999776482582, -0.00800000037997961, 0.00800000037997961, 0.008999999612569809, 0.008999999612569809, -0.006000000052154064, -0.00800000037997961, 0.00800000037997961, 0.009999999776482582, 0.009999999776482582, 0.006000000052154064, 0.00800000037997961, -0.00800000037997961, -0.00800000037997961, -0.009999999776482582, 0.008999999612569809, 0.006000000052154064, 0.009999999776482582, -0.009999999776482582, -0.00800000037997961, 0.007000000216066837, 0.008999999612569809, -0.00800000037997961, 0.009999999776482582, 0.009999999776482582, -0.008999999612569809, 0.00800000037997961]... [Tensor of shape torch.Size([1, 135])]
base_model.linear.bias: tensor([0.])

%==================================== 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([[ 0.1953],
        [-0.0139],
        [ 0.0094],
        [-0.0168],
        [ 0.0056],
        [ 0.0452],
        [ 0.0059],
        [ 0.0187],
        [ 0.1456],
        [-0.0608]], 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: coefficients.csv