DepressionBarbu#
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.DepressionBarbu)
class DepressionBarbu(pyagingModel):
def __init__(self):
super().__init__()
def preprocess(self, x):
return x
def postprocess(self, x):
return x
[3]:
model = pya.models.DepressionBarbu()
Define clock metadata#
[4]:
model.metadata["clock_name"] = "depressionbarbu"
model.metadata["data_type"] = "DNA methylation" # Paper: The model is based on DNA methylation measurements.
model.metadata["species"] = "Homo sapiens" # Paper: The study samples are Homo sapiens.
model.metadata["year"] = 2021
model.metadata["approved_by_author"] = "⌛"
model.metadata["citation"] = "Barbu, Miruna C., et al. \"Epigenetic prediction of major depressive disorder.\" Molecular Psychiatry 26.9 (2021): 5112-5123."
model.metadata["doi"] = "https://doi.org/10.1038/s41380-020-0808-3"
model.metadata["notes"] = "Blood methylation risk score for major depressive disorder built with penalised regression on genome-wide EPIC-array CpGs, trained on over 1,200 cases and 1,800 controls. Discriminates prevalent from incident MDD independently of polygenic risk, with a smoking-independent variant also derived."
model.metadata["research_only"] = None
model.metadata["tissue"] = ["whole blood"] # Paper: The listed tissue is the model-development sample material.
model.metadata["predicts"] = ["major depressive disorder"] # Paper: The reported predictor output is major depressive disorder (MDD) status/risk.
model.metadata["training_target"] = ["major depressive disorder"] # Paper: The fitting outcome is major depressive disorder (MDD) status/risk.
model.metadata["unit"] = ["unitless"] # Paper: The returned construct is expressed as score (arbitrary).
model.metadata["model_type"] = "elastic net regression" # Paper: The clock was fitted using Elastic net.
model.metadata["platform"] = ["Illumina EPIC"] # Paper: Training/selection used Illumina EPIC.
model.metadata["population"] = "adults" # Paper: adults (Generation Scotland cohort)
model.metadata["journal"] = "Molecular Psychiatry"
model.metadata["last_author"] = "Andrew M. McIntosh"
model.metadata["n_features"] = 196
model.metadata["citations"] = 93
model.metadata["citations_date"] = "2026-07-05"
Download clock dependencies#
[5]:
os.system(f"curl -sL -o depressionbarbu.xlsx https://static-content.springer.com/esm/art%3A10.1038%2Fs41380-020-0808-3/MediaObjects/41380_2020_808_MOESM4_ESM.xlsx")
[5]:
0
Load features#
[6]:
df = pd.read_excel('depressionbarbu.xlsx', sheet_name='MRS - CpG sites and weights')
model.features = df['CpG site'].tolist()
Load weights into base model#
[7]:
weights = torch.tensor(df['Beta'].tolist()).unsqueeze(0).float()
# Methylation risk score calibration intercept
intercept = torch.tensor([12.2169841]).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': 'Barbu, Miruna C., et al. "Epigenetic prediction of major '
'depressive disorder." Molecular Psychiatry 26.9 (2021): '
'5112-5123.',
'clock_name': 'depressionbarbu',
'data_type': 'methylation',
'doi': 'https://doi.org/10.1038/s41380-020-0808-3',
'notes': None,
'research_only': None,
'species': 'Homo sapiens',
'version': None,
'year': 2021}
reference_values: None
preprocess_name: None
preprocess_dependencies: None
postprocess_name: None
postprocess_dependencies: None
features: ['cg02115394', 'cg15971980', 'cg03054303', 'cg09791621', 'cg12736206', 'cg22225420', 'cg22407822', 'cg02998018', 'cg07459409', 'cg20496693', 'cg15251779', 'cg03070922', 'cg04142555', 'cg18090197', 'cg14728380', 'cg09929180', 'cg13529291', 'cg19474047', 'cg04191989', 'cg03607825', 'cg05575921', 'cg03307749', 'cg12832498', 'cg10623600', 'cg15847996', 'cg25242471', 'cg10928544', 'cg02636222', 'cg08464831', 'cg15586193']... [Total elements: 196]
base_model_features: None
%==================================== Model Details ====================================%
Model Structure:
base_model: LinearModel(
(linear): Linear(in_features=196, out_features=1, bias=True)
)
%==================================== Model Details ====================================%
Model Parameters and Weights:
base_model.linear.weight: [0.13174669444561005, 0.017368832603096962, 0.07059700042009354, -0.018956167623400688, 0.35600546002388, 0.2889443039894104, 0.0505455806851387, -0.4532535970211029, -0.2310783565044403, 0.006119664758443832, -0.07810869812965393, 0.0040334537625312805, 0.005831866059452295, 0.21422632038593292, 0.07561899721622467, 0.13462267816066742, 0.37971192598342896, -0.21138310432434082, 0.1820078045129776, 0.06565909832715988, -0.14514309167861938, -0.3468712866306305, 0.13083887100219727, 0.03079209290444851, -0.06906116753816605, 0.11012744158506393, 0.01911790855228901, 0.5892500281333923, -0.025107210502028465, -0.22120502591133118]... [Tensor of shape torch.Size([1, 196])]
base_model.linear.bias: tensor([12.2170])
%==================================== 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([[11.7817],
[12.6549],
[11.4798],
[12.3791],
[11.4735],
[ 7.2415],
[18.4420],
[14.8113],
[14.2788],
[13.7944]], 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: depressionbarbu.xlsx