Lin#
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.Lin)
class Lin(pyagingModel):
def __init__(self):
super().__init__()
def preprocess(self, x):
return x
def postprocess(self, x):
return x
[3]:
model = pya.models.Lin()
Define clock metadata#
[4]:
model.metadata["clock_name"] = 'lin'
model.metadata["data_type"] = 'methylation'
model.metadata["species"] = 'Homo sapiens'
model.metadata["year"] = 2016
model.metadata["approved_by_author"] = '⌛'
model.metadata["citation"] = "Lin, Qiong, et al. \"DNA methylation levels at individual age-associated CpG sites can be indicative for life expectancy.\" Aging (Albany NY) 8.2 (2016): 394."
model.metadata["doi"] = "https://doi.org/10.18632/aging.100908"
model.metadata["research_only"] = None
model.metadata["notes"] = None
Download clock dependencies#
[5]:
# from https://www.aging-us.com/article/100908/text supplement
cpg_sites = [
"(Intercept)", "cg05228408", "cg16352283", "cg05436231", "cg19046959",
"cg17791651", "cg07388493", "cg04036898", "cg07810156", "cg21448423",
"cg18660898", "cg25256723", "cg21870884", "cg25947945", "cg09462576",
"cg09809672", "cg27553955", "cg27320127", "cg15297650", "cg05331214",
"cg24178740", "cg18182399", "cg25431974", "cg24768561", "cg26614073",
"cg23320649", "cg12554573", "cg04474832", "cg17421623", "cg22919728",
"cg14456683", "cg08209133", "cg16744741", "cg00059225", "cg00489401",
"cg02844545", "cg22736354", "cg06493994", "cg03340878", "cg03958979",
"cg15804973", "cg13870866", "cg00503840", "cg25762706", "cg25538571",
"cg08598221", "cg19724470", "cg07211259", "cg13870494", "cg16386080",
"cg00563932", "cg21120249", "cg26581729", "cg17431739", "cg13129046",
"cg01560871", "cg06291867", "cg26610808", "cg07621046", "cg13807496",
"cg20654468", "cg21992250", "cg15538427", "cg08012287", "cg01820374",
"cg19722847", "cg12883767", "cg04123409", "cg22580512", "cg25268718",
"cg21296230", "cg21801378", "cg10917602", "cg15195412", "cg20264732",
"cg22947000", "cg02228185", "cg01739167", "cg14918082", "cg05379350",
"cg08468689", "cg08090640", "cg25809905", "cg05294455", "cg06638433",
"cg20366832", "cg19761273", "cg26927807", "cg17471102", "cg02489552",
"cg05488632", "cg16363586", "cg17861230", "cg24713204", "cg23679724",
"cg03224418", "cg15379633", "cg02994956", "cg23124451", "cg26394940"
]
coefficients = [
12.2169841, 0.47636173, -5.3124138, 17.7305146, -13.367066, 8.72680959, -4.7759575,
10.162153, 15.3892025, -4.4621797, 13.2543665, -11.802998, 22.9981412, -8.2387336,
6.3124836, -14.950409, -0.7884001, 48.9368049, -34.306553, 9.83640629, -27.476107,
-4.1527608, -1.048605, -4.5917403, -11.443446, 8.70555476, 1.81880164, -26.556597,
2.4399993, 0.99214006, 13.1457167, 30.500322, -9.5846721, 36.8195086, 1.98682848,
0.38022482, 36.9317174, 66.1611861, 5.95485236, -16.016804, -15.214138, -39.104364,
31.2113275, 1.5340163, 10.2956593, 2.62080161, -5.5537073, -12.424324, 19.7417678,
-29.24993, -3.5009711, -8.6074197, -7.9914389, 8.22589722, -5.1368284, 13.5034883,
13.0769424, -21.374356, 13.6468199, -8.3931276, 8.14605552, 10.5216611, -19.93487,
-18.989957, -30.896866, -13.06341, 8.45912249, -10.767354, -0.8383178, 4.06576438,
28.1787443, 44.7163476, -6.0293979, 20.050343, -20.618882, -13.217155, -8.6363427,
33.8101434, 15.5554908, 17.340667, -16.062905, 8.31318309, -6.0974732, 2.71073045,
10.6229217, 2.97899616, -16.331359, 16.5195276, -18.063487, 6.09699424, -11.249025,
13.6452671, 17.5027126, -32.487323, 0.25793126, 8.07556639, 15.4139903, -6.4516149,
-13.361462, 0.89292205
]
Load features#
[6]:
df = pd.DataFrame({
'feature': cpg_sites,
'coefficient': coefficients
})
model.features = features = df['feature'][1:].tolist()
Load weights into base model#
[7]:
weights = torch.tensor(df['coefficient'][1:].tolist()).unsqueeze(0)
intercept = torch.tensor([df['coefficient'][0]])
Linear model#
[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': 'Lin, Qiong, et al. "DNA methylation levels at individual '
'age-associated CpG sites can be indicative for life expectancy." '
'Aging (Albany NY) 8.2 (2016): 394.',
'clock_name': 'lin',
'data_type': 'methylation',
'doi': 'https://doi.org/10.18632/aging.100908',
'notes': None,
'research_only': None,
'species': 'Homo sapiens',
'version': None,
'year': 2016}
reference_values: None
preprocess_name: None
preprocess_dependencies: None
postprocess_name: None
postprocess_dependencies: None
features: ['cg05228408', 'cg16352283', 'cg05436231', 'cg19046959', 'cg17791651', 'cg07388493', 'cg04036898', 'cg07810156', 'cg21448423', 'cg18660898', 'cg25256723', 'cg21870884', 'cg25947945', 'cg09462576', 'cg09809672', 'cg27553955', 'cg27320127', 'cg15297650', 'cg05331214', 'cg24178740', 'cg18182399', 'cg25431974', 'cg24768561', 'cg26614073', 'cg23320649', 'cg12554573', 'cg04474832', 'cg17421623', 'cg22919728', 'cg14456683']... [Total elements: 99]
base_model_features: None
%==================================== Model Details ====================================%
Model Structure:
base_model: LinearModel(
(linear): Linear(in_features=99, out_features=1, bias=True)
)
%==================================== Model Details ====================================%
Model Parameters and Weights:
base_model.linear.weight: [0.4763617217540741, -5.312413692474365, 17.730514526367188, -13.367066383361816, 8.72680950164795, -4.7759575843811035, 10.162153244018555, 15.389202117919922, -4.462179660797119, 13.254366874694824, -11.802997589111328, 22.998140335083008, -8.238733291625977, 6.312483787536621, -14.950408935546875, -0.7884001135826111, 48.936805725097656, -34.30655288696289, 9.836406707763672, -27.476106643676758, -4.152760982513428, -1.048604965209961, -4.591740131378174, -11.443446159362793, 8.705554962158203, 1.8188016414642334, -26.556596755981445, 2.4399993419647217, 0.9921400547027588, 13.145716667175293]... [Tensor of shape torch.Size([1, 99])]
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([[ 305.0711],
[ 247.5204],
[ 118.7530],
[ 22.4327],
[ -22.3631],
[ 186.0485],
[-195.4503],
[ 222.6908],
[ 25.2435],
[ 281.9471]], 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)