### Install Biolearn using pip

Source: https://bio-learn.github.io/quickstart.html

Use this command to install the biolearn library. Ensure you have Python 3.10+ installed.

```bash
pip install biolearn

```

--------------------------------

### Deconvolution Example

Source: https://bio-learn.github.io/auto_examples/index.html

This example shows how to perform deconvolution on a dataset with known cell proportions. Ensure you have write permissions in the directory where you run the examples.

```python
import biolearn

# Example of deconvolution
# ... (actual code would go here)
```

--------------------------------

### Exploring the Challenge Data

Source: https://bio-learn.github.io/auto_examples/index.html

This example shows how to explore the data provided for a challenge. Ensure you have write permissions in the directory where you run the examples.

```python
import biolearn

# Example of exploring challenge data
# ... (actual code would go here)
```

--------------------------------

### Training an ElasticNet Model

Source: https://bio-learn.github.io/auto_examples/index.html

This example demonstrates how to train an ElasticNet model. Ensure you have write permissions in the directory where you run the examples.

```python
import biolearn

# Example of training an ElasticNet model
# ... (actual code would go here)
```

--------------------------------

### Building a Competition Submission

Source: https://bio-learn.github.io/auto_examples/index.html

This example demonstrates how to build a competition submission using an existing model. Ensure you have write permissions in the directory where you run the examples.

```python
import biolearn

# Example of building a competition submission
# ... (actual code would go here)
```

--------------------------------

### Hurdle InflammAge API Example

Source: https://bio-learn.github.io/auto_examples/index.html

This example demonstrates how to use the Hurdle InflammAge API. Ensure you have write permissions in the directory where you run the examples.

```python
import biolearn

# Example usage of Hurdle InflammAge API
# ... (actual code would go here)
```

--------------------------------

### Local Data Loading Example

Source: https://bio-learn.github.io/auto_examples/index.html

This example shows how to load data locally using Biolearn. Ensure you have write permissions in the directory where you run the examples.

```python
import biolearn

# Example of local data loading
# ... (actual code would go here)
```

--------------------------------

### Clock/Model Visualizations using GEO Datasets

Source: https://bio-learn.github.io/auto_examples/index.html

This example shows how to visualize clocks and models using GEO datasets. Ensure you have write permissions in the directory where you run the examples.

```python
import biolearn

# Example of clock/model visualizations with GEO datasets
# ... (actual code would go here)
```

--------------------------------

### Verify Biolearn Installation

Source: https://bio-learn.github.io/quickstart.html

After installation, run this code in a Python interpreter or Jupyter notebook to confirm that the DataLibrary class can be imported without errors.

```python
from biolearn.data_library import DataLibrary

```

--------------------------------

### Quality Control Visualization using GEO Datasets

Source: https://bio-learn.github.io/auto_examples/index.html

This example demonstrates visualization for quality control using GEO datasets. Ensure you have write permissions in the directory where you run the examples.

```python
import biolearn

# Example of quality control visualization with GEO datasets
# ... (actual code would go here)
```

--------------------------------

### DNA Methylation Visualizations using GEO Datasets

Source: https://bio-learn.github.io/auto_examples/index.html

This example demonstrates DNA methylation visualizations using GEO datasets. Ensure you have write permissions in the directory where you run the examples.

```python
import biolearn

# Example of DNA methylation visualizations with GEO datasets
# ... (actual code would go here)
```

--------------------------------

### Transcriptomic Clock using GEO Data

Source: https://bio-learn.github.io/auto_examples/index.html

This example demonstrates how to use a transcriptomic clock with data from GEO. Ensure you have write permissions in the directory where you run the examples.

```python
import biolearn

# Example of transcriptomic clock using GEO data
# ... (actual code would go here)
```

--------------------------------

### Epigenetic Clocks using GEO Data

Source: https://bio-learn.github.io/auto_examples/index.html

This example demonstrates how to use epigenetic clocks with data from GEO. Ensure you have write permissions in the directory where you run the examples.

```python
import biolearn

# Example of epigenetic clocks using GEO data
# ... (actual code would go here)
```

--------------------------------

### Phenotypic Ages in NHANES Data

Source: https://bio-learn.github.io/auto_examples/index.html

This example shows how to calculate and plot phenotypic ages using NHANES data. Ensure you have write permissions in the directory where you run the examples.

```python
import biolearn

# Example of phenotypic ages in NHANES data
# ... (actual code would go here)
```

--------------------------------

### Performing Custom Imputations

Source: https://bio-learn.github.io/auto_examples/index.html

This example demonstrates how to perform custom data imputations using Biolearn. Ensure you have write permissions in the directory where you run the examples.

```python
import biolearn

# Example of performing custom imputations
# ... (actual code would go here)
```

--------------------------------

### Down Syndrome Epigenetic Plotting

Source: https://bio-learn.github.io/auto_examples/index.html

This example shows how to perform epigenetic plotting for Down Syndrome data. Ensure you have write permissions in the directory where you run the examples.

```python
import biolearn

# Example of Down Syndrome epigenetic plotting
# ... (actual code would go here)
```

--------------------------------

### Import BioLearn Classes and Functions

Source: https://bio-learn.github.io/_downloads/37a16eeb64cb047f7329c72df903f23a/plot_quality_control.ipynb

Imports DataLibrary, GeoData, and plot_sample_deviations from the biolearn library. Ensure these modules are installed.

```python
from biolearn.data_library import DataLibrary, GeoData
from biolearn.visualize import plot_sample_deviations
```

--------------------------------

### Load NHANES 2010 Data

Source: https://bio-learn.github.io/auto_examples/01_composite_biomarkers/plot_nhanes.html

Loads blood exam data from NHANES 2010 and calculates 'years_until_death'. Ensure the biolearn library is installed.

```python
from biolearn.load import load_nhanes
year = 2010
df = load_nhanes(year)
df["years_until_death"] = df["months_until_death"] / 12
```

--------------------------------

### Regression statistics output

Source: https://bio-learn.github.io/auto_examples/04_additional_visualizations/plot_dnam_relationships.html

Example output showing R-squared and Pearson correlation values for a specific CpG site.

```text
Computed statistics for CpG site: cg00000029
Linear Regression R^2: 0.0311
Pearson Correlation (r): -0.1764
Pearson Correlation p-value: 8.2813e-03
Polynomial Regression (Degree 2) R^2: 0.0327
```

--------------------------------

### Metadata CSV Format

Source: https://bio-learn.github.io/methylation-standard.html

Example of a metadata file formatted as a CSV with sample identifiers and associated clinical fields.

```text
SampleID    Sex    Age     Disease State
GSM3074480    0    25.2    None
GSM3074481    0    29.5    COVID
GSM3074482    1    39.7    None
```

--------------------------------

### GET biolearn.load.load_nhanes

Source: https://bio-learn.github.io/modules/generated/biolearn.load.load_nhanes.html

Loads data from the National Health and Nutrition Examination Survey (NHANES) for a specified year.

```APIDOC
## GET biolearn.load.load_nhanes

### Description
Loads data from the National Health and Nutrition Examination Survey (NHANES). Note that NHANES data comes in two-year groupings and the year passed in should be the later year.

### Parameters
#### Query Parameters
- **year** (number) - Required - A year number for which to load data. Supported inputs are 2010 and 2012.

### Response
#### Success Response (200)
- **df** (Pandas.Dataframe) - A pandas dataframe where each row represents an individual and each column represents a measurement about that individual.
```

--------------------------------

### DNA Methylation Data Format

Source: https://bio-learn.github.io/methylation-standard.html

Example of a DNA methylation file showing CpG sites as rows and samples as columns with beta values.

```text
cpgSite     GSM3074480    GSM3074481    GSM3074482    GSM3074483
cg00000029    0.395         0.499         0.42         0.423
cg00000109    0.821         0.868         0.877        0.841
cg00000155    0.878         0.89          0.894        0.822
```

--------------------------------

### GET biolearn.cache.NoCache.get

Source: https://bio-learn.github.io/modules/generated/biolearn.cache.NoCache.html

Retrieves a cache entry. In the NoCache implementation, this method always returns None, indicating a cache miss.

```APIDOC
## GET biolearn.cache.NoCache.get

### Description
Retrieve a cache entry. This method always returns None, indicating a cache miss.

### Parameters
#### Query Parameters
- **key** (str) - Required - Unique key identifying the cache entry.
- **category** (str) - Required - Cache category name.
- **version** (str) - Required - Version string for invalidation.

### Response
#### Success Response (200)
- **result** (None) - Always returns None.
```

--------------------------------

### LocalFolderCache Initialization

Source: https://bio-learn.github.io/modules/generated/biolearn.cache.LocalFolderCache.html

Initializes the LocalFolderCache with a specified path and maximum size.

```APIDOC
## LocalFolderCache.__init__

### Description
Initialize the LocalFolderCache.

### Method
__init__

### Parameters
#### Path Parameters
- **path** (str) - Required - Directory path for storing cache files.
- **max_size_gb** (float) - Required - Maximum total cache size in gigabytes.
```

--------------------------------

### Import biolearn visualization classes and functions

Source: https://bio-learn.github.io/_downloads/108af8868614eea84b27d2927b9893eb/plot_clock_model_relationships.ipynb

Initializes the necessary modules for data loading, model retrieval, and visualization.

```python
from biolearn.data_library import DataLibrary
from biolearn.model_gallery import ModelGallery
from biolearn.visualize import (
    plot_clock_correlation_matrix,
    plot_clock_deviation_heatmap,
    plot_age_prediction,
    plot_health_outcome
)
```

--------------------------------

### ModelGallery Initialization

Source: https://bio-learn.github.io/modules/generated/biolearn.model_gallery.ModelGallery.html

Initializes the ModelGallery instance with a dictionary of model definitions.

```APIDOC
## ModelGallery.__init__

### Description
Initializes the ModelGallery instance.

### Method
__init__

### Parameters
#### Request Body
- **models** (dict) - Required - A dictionary of model definitions.

### Request Example
```json
{
  "models": {
    "model1": { ... },
    "model2": { ... }
  }
}
```

### Response
#### Success Response (200)
- **ModelGallery instance** (object) - The initialized ModelGallery object.

### Response Example
```json
{
  "message": "ModelGallery initialized successfully"
}
```
```

--------------------------------

### LocalFolderCache Initialization

Source: https://bio-learn.github.io/data.html

Initialize `LocalFolderCache` to use a local folder for caching. Specify the directory path for storing cached data.

```python
from biolearn.cache import LocalFolderCache

cache = LocalFolderCache(cache_dir='/path/to/cache')
```

--------------------------------

### DataLibrary Get Source

Source: https://bio-learn.github.io/modules/generated/biolearn.data_library.DataLibrary.html

Retrieves a data source by its identifier.

```APIDOC
## DataLibrary.get

### Description
Retrieves a data source by its identifier.

### Method
get

### Parameters
#### Path Parameters
- **source_id** (str) - Required - The identifier of the data source to retrieve.

### Response
#### Success Response (200)
- **source** (object) - The data source with the given identifier if found, otherwise None.
```

--------------------------------

### GET biolearn.load.load_fhs

Source: https://bio-learn.github.io/modules/generated/biolearn.load.load_fhs.html

Retrieves the Framingham Heart Study dataset as a pandas DataFrame.

```APIDOC
## biolearn.load.load_fhs

### Description
Loads data from the Framingham Heart Study.

### Response
#### Success Response (200)
- **df** (Pandas.Dataframe) - A pandas dataframe where each row represents an individual and each column represents a measurement about that individual.
```

--------------------------------

### DataLibrary Initialization

Source: https://bio-learn.github.io/modules/generated/biolearn.data_library.DataLibrary.html

Initializes the DataLibrary instance with an optional library file and cache mechanism.

```APIDOC
## DataLibrary.__init__

### Description
Initializes the DataLibrary instance with an optional library file and cache mechanism.

### Method
__init__

### Parameters
#### Path Parameters
- **library_file** (str) - Optional - The path to the library file. If None, the default biolearn library file is loaded.
- **cache** (object) - Optional - An object that adheres to the caching interface used in the caching module. If None, the default cache is used. This cache will be used by all returned data sources
```

--------------------------------

### DataLibrary Initialization

Source: https://bio-learn.github.io/clocks.html

Initialize a `DataLibrary` object from `biolearn.data_library` to manage data sources. Specify the directory where data is stored.

```python
from biolearn.data_library import DataLibrary

data_library = DataLibrary(data_dir="/path/to/data")
```

--------------------------------

### Split data into training and test sets

Source: https://bio-learn.github.io/_downloads/9d6419c1b7da427cabe6da2ae524be9e/training_simple_model.ipynb

Prepares the feature matrix and target vector, then splits the data for model evaluation.

```python
from sklearn.model_selection import train_test_split

df = data.dnam.transpose()
df['age'] = data.metadata['age']
top_sites_df = df[stable_cpg_names]

X = top_sites_df
y = df['age']

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
```

--------------------------------

### DataSource Initialization

Source: https://bio-learn.github.io/clocks.html

Create a `DataSource` object from `biolearn.data_library` to represent a specific data source. This is typically done within a `DataLibrary` context.

```python
from biolearn.data_library import DataSource

source = DataSource(name="my_source", data_dir="/path/to/source")
```

--------------------------------

### Load Data and Initialize Hurdle InflammAge API

Source: https://bio-learn.github.io/auto_examples/hurdle_api_example.html

Loads sample data from the DataLibrary and checks for the required HURDLE_API_KEY environment variable before proceeding with model operations.

```python
import os
import pandas as pd
from biolearn.model_gallery import ModelGallery
from biolearn.data_library import DataLibrary

print("Loading sample data...")
data = DataLibrary().get("BoAChallengeData").load()

# Use first 5 samples for quick testing
sample_data = data.dnam.iloc[:, :5]
print(f"Using {sample_data.shape[1]} samples with {sample_data.shape[0]} CpG sites")

gallery = ModelGallery()

# Check if API key is set
if not os.environ.get("HURDLE_API_KEY"):
    print("\nNo HURDLE_API_KEY environment variable found.")
    api_key = input("Enter your Hurdle API key: ").strip()
    if api_key:
        os.environ["HURDLE_API_KEY"] = api_key
    else:
        print("No API key provided. Exiting.")
        exit(1)
```

--------------------------------

### ModelGallery Initialization

Source: https://bio-learn.github.io/data.html

Initialize `ModelGallery` to access a collection of pre-trained models. This class is central to applying existing models to new data.

```python
from biolearn.model_gallery import ModelGallery

model_gallery = ModelGallery()
```

--------------------------------

### Load NHANES 2010 Data

Source: https://bio-learn.github.io/_downloads/6ba1afcbb57f61d1d68110657ea7d4a7/plot_nhanes.ipynb

Imports the NHANES dataset and calculates the years until death from the provided months.

```python
from biolearn.load import load_nhanes 
year = 2010
df = load_nhanes(year)
df["years_until_death"] = df["months_until_death"] / 12
```

--------------------------------

### Load Methylation Data from GEO

Source: https://bio-learn.github.io/auto_examples/00_omics_biomarkers/plot_down_syndrome_model.html

Loads DNA methylation data from a specified GEO accession using the DataLibrary. Ensure the 'biolearn' library is installed.

```python
from biolearn.data_library import DataLibrary
test_data = DataLibrary().get("GSE52588").load()
```

--------------------------------

### Generate Quality Reports for Datasets

Source: https://bio-learn.github.io/_downloads/37a16eeb64cb047f7329c72df903f23a/plot_quality_control.ipynb

Iterates through the loaded datasets and generates a quality report for each. The .show() method displays the report.

```python
[dataset.quality_report().show() for dataset in datasets.values()]
```

--------------------------------

### Load GEO Blood Methylation Dataset

Source: https://bio-learn.github.io/_downloads/ed93f9c54e6f01466b8a5a0db47f22b0/plot_blood_deconvolution.ipynb

Loads the GSE112618 dataset, which contains blood DNA methylation data measured on the EPIC platform. Ensure the biolearn library is installed.

```python
from biolearn.data_library import DataLibrary

epic_facs_salas_18 = DataLibrary().get("GSE112618").load()
```

--------------------------------

### Plot Predicted vs. Chronological Age

Source: https://bio-learn.github.io/auto_examples/00_omics_biomarkers/plot_transcriptomic_clock.html

Calculates the Pearson correlation coefficient and generates a scatter plot comparing predicted ages with chronological ages using seaborn and matplotlib. Requires 'seaborn', 'matplotlib', and 'scipy' to be installed.

```python
import matplotlib.pyplot as plt
import seaborn as sns
from scipy.stats import pearsonr

predicted_values = predicted['Predicted']
true_values = data.metadata['age']

correlation, _ = pearsonr(true_values, predicted_values)
print(f"Pearson Correlation Coefficient: {correlation}")

sns.set_theme(style="whitegrid")
sns.regplot(x=true_values, y=predicted_values, scatter_kws={'alpha':0.5})
plt.title('Transcriptomic Prediction Model')
plt.xlabel('Chronological Age')
plt.ylabel('Transcriptomic Age')
plt.show()
```

```text
Pearson Correlation Coefficient: 0.7025211774998293
```

--------------------------------

### Load Challenge Data with BioLearn

Source: https://bio-learn.github.io/_downloads/e2c40c72c94a18149a91c1d47ee4de74/plot_challenge_data.ipynb

Loads the challenge dataset using DataLibrary. Use this to access competition data.

```python
from biolearn.data_library import DataLibrary

challenge_data = DataLibrary().get("BoAChallengeData").load()
```

--------------------------------

### DataLibrary Initialization

Source: https://bio-learn.github.io/data.html

Initialize `DataLibrary` to manage data sources. This is a core component for accessing various datasets within Biolearn.

```python
from biolearn.data_library import DataLibrary

data_library = DataLibrary()
```

--------------------------------

### Process Challenge Data for Prediction

Source: https://bio-learn.github.io/_downloads/9d6419c1b7da427cabe6da2ae524be9e/training_simple_model.ipynb

Prepares challenge data by transposing, selecting stable CpG sites, filling missing values with 0, and then predicting ages using the trained model.

```python
pruned_data = challenge_data.dnam.T[stable_cpg_names]
pruned_data = pruned_data.fillna(0)
challenge_results = model.predict(pruned_data)
```

--------------------------------

### NoCache Initialization

Source: https://bio-learn.github.io/data.html

Initialize `NoCache` to disable caching mechanisms. This is useful when explicit control over caching is required.

```python
from biolearn.cache import NoCache

cache = NoCache()
```

--------------------------------

### DataSource Initialization

Source: https://bio-learn.github.io/data.html

Initialize `DataSource` to represent a specific data source. This class is used in conjunction with `DataLibrary`.

```python
from biolearn.data_library import DataSource

my_source = DataSource(name='my_data', path='/path/to/data')
```

--------------------------------

### Import biolearn visualization modules

Source: https://bio-learn.github.io/_downloads/a78e2414f22fb790bca7559350d8cb52/plot_dnam_relationships.ipynb

Import necessary classes and functions from the biolearn library to perform methylation analysis.

```python
from biolearn.data_library import DataLibrary
from biolearn.visualize import (
    identify_stable_cpg_sites,
    compute_age_group_stats,
    compute_methylation_stats,
    plot_methylation_by_age_sex,
    plot_methylation_vs_age
)
```

--------------------------------

### GeoData Initialization

Source: https://bio-learn.github.io/clocks.html

Initialize a `GeoData` object from `biolearn.data_library` to handle GEO datasets. Provide the path to the GEO data.

```python
from biolearn.data_library import GeoData

geo_data = GeoData(data_dir="/path/to/geo/data")
```

--------------------------------

### Data Loading Utilities

Source: https://bio-learn.github.io/modules/load.html

Functions for loading datasets.

```APIDOC
## `biolearn.load`: Data Loading Utilities

### Description
Provides functions to load datasets from various sources.

### Functions

#### `load_nhanes(year)`

##### Description
Loads data from the National Health and Nutrition Examination Survey.

##### Parameters
- **year** (int) - Required - The year of the NHANES data to load.

##### Response Example
```json
{
  "data": "NHANES dataset for the specified year"
}
```

#### `load_fhs()`

##### Description
Loads data from the Framingham Heart Study.

##### Parameters
None

##### Response Example
```json
{
  "data": "Framingham Heart Study dataset"
}
```
```

--------------------------------

### Load Competition Data with GeoData

Source: https://bio-learn.github.io/_downloads/097efd573efd69c248c612b7d5ada8b7/simple_competiton_submission.ipynb

Loads methylation data for a competition using the GeoData class. Ensure you replace 'ADD YOUR PATH HERE' with the actual path to your downloaded data file.

```python
from biolearn.data_library import GeoData

#Download the data file for the warmup challenge linked here https://www.synapse.org/#!Synapse:syn52966292/wiki/625231
DOWNLOADED_DATA_FILE_PATH="ADD YOUR PATH HERE"
challenge_data = GeoData.from_methylation_matrix(DOWNLOADED_DATA_FILE_PATH)
challenge_data
```

--------------------------------

### Load GEO methylation datasets

Source: https://bio-learn.github.io/_downloads/a78e2414f22fb790bca7559350d8cb52/plot_dnam_relationships.ipynb

Initialize the DataLibrary and load specific GEO datasets into a dictionary of GeoData objects.

```python
# List dataset IDs
dataset_ids = ["GSE112618", "GSE110554", "GSE41169", "GSE52588"]

# Load datasets into a dictionary of GeoData objects
library = DataLibrary()
datasets = {id: library.get(id).load() for id in dataset_ids}
```

--------------------------------

### LocalFolderCache.get

Source: https://bio-learn.github.io/modules/generated/biolearn.cache.LocalFolderCache.html

Retrieves a cached value using its key, category, and version.

```APIDOC
## LocalFolderCache.get

### Description
Retrieve a cached value.

### Method
get

### Parameters
#### Path Parameters
- **key** (str) - Required - Unique key identifying the cache entry.
- **category** (str) - Required - Cache category name.
- **version** (str) - Required - Version string for invalidation.

### Returns
Any | None - The cached object, or None if not found or on load error.
```

--------------------------------

### biolearn.load.load_nhanes

Source: https://bio-learn.github.io/modules/generated/biolearn.imputation.impute_from_average.html

Loads data from the NHANES dataset.

```APIDOC
## load_nhanes

### Description
Loads data from the NHANES dataset.

### Method
N/A (Function Signature)

### Endpoint
N/A (Function Signature)

### Parameters
#### Path Parameters
N/A

#### Query Parameters
N/A

#### Request Body
N/A

### Request Example
N/A

### Response
#### Success Response (200)
- **data** (pd.DataFrame) - DataFrame containing the loaded NHANES data.

#### Response Example
N/A
```

--------------------------------

### biolearn.model_gallery : ModelGallery

Source: https://bio-learn.github.io/modules/generated/biolearn.model_gallery.ModelGallery.html

Reference for the ModelGallery class.

```APIDOC
## biolearn.model_gallery.ModelGallery

### Description
Provides access to a gallery of pre-trained models.

### Parameters
This class does not have explicit parameters listed in the provided text.
```

--------------------------------

### GeoData Class Initialization

Source: https://bio-learn.github.io/modules/generated/biolearn.data_library.GeoData.html

Initializes a GeoData instance with metadata and optional methylation data.

```APIDOC
## GeoData Class

### Description
Represents genomic data with a focus on metadata and methylation data. GeoData facilitates the organization and access to metadata and methylation data.

### Parameters
- **metadata** (DataFrame) - Required - Metadata associated with genomic samples.
- **dnam** (DataFrame) - Optional - Methylation data associated with genomic samples.
- **rna** (DataFrame) - Optional - RNA data associated with genomic samples.
- **protein_alamar** (DataFrame) - Optional - Protein data associated with genomic samples.
- **protein_olink** (DataFrame) - Optional - Protein data associated with genomic samples.
```

--------------------------------

### Load GEO Datasets into DataLibrary

Source: https://bio-learn.github.io/_downloads/37a16eeb64cb047f7329c72df903f23a/plot_quality_control.ipynb

Creates a DataLibrary instance and loads specified GEO datasets. The datasets are stored in a dictionary mapping dataset IDs to loaded GeoData objects.

```python
library = DataLibrary()
dataset_ids = ["GSE112618", "GSE110554", "GSE41169", "GSE52588"]
datasets = {id: library.get(id).load() for id in dataset_ids}
```

--------------------------------

### Search for datasets using DataLibrary

Source: https://bio-learn.github.io/modules/generated/biolearn.data_library.DataLibrary.html

Demonstrates filtering datasets by metadata criteria and retrieving available metadata fields.

```python
>>> # Find all datasets with female subjects
>>> library = DataLibrary()
>>> female_datasets = library.search(sex="female")
```

```python
>>> # Find datasets with elderly subjects (70+ years)
>>> elderly_datasets = library.search(min_age=70)
```

```python
>>> # Find male datasets with subjects over 50
>>> male_elderly = library.search(sex="male", min_age=50)
```

```python
>>> # View available metadata fields
>>> all_datasets = library.search()
>>> print(all_datasets.columns.tolist())
```

--------------------------------

### Data Loading Utilities

Source: https://bio-learn.github.io/index.html

Utilities for loading data from various sources.

```APIDOC
## Data Loading Utilities

### Description
Provides functions to load data from common sources like NHANES, FHS, and GEO.

### Functions

#### `biolearn.load.load_nhanes`

Loads data from the National Health and Nutrition Examination Survey (NHANES).

### Parameters

(Specific parameters not detailed in the provided text)

### Request Example

(Not available in the provided text)

### Response

(Not available in the provided text)

#### `biolearn.load.load_fhs`

Loads data from the Framingham Heart Study (FHS).

### Parameters

(Specific parameters not detailed in the provided text)

### Request Example

(Not available in the provided text)

### Response

(Not available in the provided text)
```

--------------------------------

### biolearn.load : Data Loading Utilities

Source: https://bio-learn.github.io/modules/generated/biolearn.model_gallery.ModelGallery.html

Reference for data loading utility functions.

```APIDOC
## biolearn.load.load_nhanes

### Description
Loads data from the NHANES dataset.

### Parameters
This function does not have explicit parameters listed in the provided text.

## biolearn.load.load_fhs

### Description
Loads data from the FHS dataset.

### Parameters
This function does not have explicit parameters listed in the provided text.
```

--------------------------------

### Load Model and Predict InflammAge

Source: https://bio-learn.github.io/auto_examples/hurdle_api_example.html

Loads the Hurdle InflammAge model and makes predictions on sample data. Requires user consent for data transmission. Handles potential API errors.

```python
model = gallery.get("HurdleInflammAge")

print("\nMaking predictions...")
print("Note: You will be prompted to consent to sending data to Hurdle's servers.")

try:
    predictions = model.predict(sample_data)

    print("\nInflammAge Results:")
    print("-" * 40)
    for sample, age in predictions.items():
        print(f"{sample}: {age:.1f} years")

    # If metadata is available, show age acceleration
    if hasattr(data, "metadata") and "age" in data.metadata.columns:
        print("\nAge Acceleration (InflammAge - Chronological Age):")
        print("-" * 40)
        for sample in predictions.index:
            if sample in data.metadata.index:
                inflamm_age = predictions[sample]
                chrono_age = data.metadata.loc[sample, "age"]
                delta = inflamm_age - chrono_age
                print(f"{sample}: {delta:+.1f} years")

except ValueError as e:
    print(f"Error: {e}")
except Exception as e:
    print(f"API Error: {e}")
    print("\nTroubleshooting:")
    print("- Verify your API key is correct")
    print("- Check your internet connection")
    print("- Ensure you have access to https://api.hurdle.bio")
```

--------------------------------

### biolearn.load: Data Loading Utilities

Source: https://bio-learn.github.io/modules/model_gallery.html

Utilities for loading various biological datasets.

```APIDOC
## `biolearn.load`: Data Loading Utilities

### Description
Utilities for loading various biological datasets.

### Functions
- **load_nhanes**: Loads data from the NHANES dataset.
- **load_fhs**: Loads data from the Framingham Heart Study (FHS) dataset.
```

--------------------------------

### biolearn.load

Source: https://bio-learn.github.io/authors.html

Utilities for loading specific biological datasets.

```APIDOC
## biolearn.load

### Description
Functions to load standardized datasets into the environment.

### Functions
- **load_nhanes**: Load NHANES dataset.
- **load_fhs**: Load FHS dataset.
```

--------------------------------

### Model Gallery

Source: https://bio-learn.github.io/index.html

Access to a gallery of pre-implemented aging models.

```APIDOC
## Model Gallery

### Description
Provides access to a gallery of pre-implemented aging models.

### Class

#### `biolearn.model_gallery.ModelGallery`

Represents the model gallery.

### Methods

(Specific methods not detailed in the provided text)

### Request Example

(Not available in the provided text)

### Response

(Not available in the provided text)
```

--------------------------------

### biolearn.load

Source: https://bio-learn.github.io/modules/generated/biolearn.imputation.hybrid_impute.html

Contains utilities for loading data, specifically supporting NHANES and FHS datasets.

```APIDOC
## biolearn.load: Data Loading Utilities

### Description
This module provides utility functions for loading specific datasets.

### Functions
- `load_nhanes`
- `load_fhs`
```

--------------------------------

### Load training data

Source: https://bio-learn.github.io/_downloads/9d6419c1b7da427cabe6da2ae524be9e/training_simple_model.ipynb

Retrieves and loads a specific dataset from the DataLibrary.

```python
from biolearn.data_library import DataLibrary
data = DataLibrary().get("GSE40279").load()
data.metadata
```

--------------------------------

### Load NHANES Data

Source: https://bio-learn.github.io/clocks.html

Use `biolearn.load.load_nhanes` to load data from the National Health and Nutrition Examination Survey. Ensure the NHANES data is downloaded and accessible.

```python
from biolearn.load import load_nhanes

data = load_nhanes(data_dir="/path/to/nhanes/data")
```

--------------------------------

### DataLibrary Load Sources

Source: https://bio-learn.github.io/modules/generated/biolearn.data_library.DataLibrary.html

Loads data sources from a given library file, appending them to the current set of data sources.

```APIDOC
## DataLibrary.load_sources

### Description
Loads data sources from a given library file appending them to the current set of data sources.

### Method
load_sources

### Parameters
#### Path Parameters
- **library_file** (str) - Required - The file path of the library file to load data sources from.
```

--------------------------------

### ModelGallery API

Source: https://bio-learn.github.io/modules/generated/biolearn.cache.LocalFolderCache.html

Reference for the ModelGallery class.

```APIDOC
## ModelGallery API

### Description
Manages a gallery of pre-trained models.

### Classes
- **biolearn.model_gallery.ModelGallery**: Class for interacting with the model gallery.
```

--------------------------------

### Visualize Age Relationship

Source: https://bio-learn.github.io/_downloads/6ba1afcbb57f61d1d68110657ea7d4a7/plot_nhanes.ipynb

Creates a scatter plot to compare chronological age against calculated phenotypic age.

```python
import seaborn as sn
sn.scatterplot(data=df,x="age", y="phenotypic_age",s=2);
```

--------------------------------

### Data Loading Utilities API

Source: https://bio-learn.github.io/modules/generated/biolearn.cache.LocalFolderCache.html

Utilities for loading specific datasets.

```APIDOC
## Data Loading Utilities API

### Description
Provides utility functions for loading specific datasets.

### Functions
- **biolearn.load.load_nhanes**: Loads data from the NHANES dataset.
- **biolearn.load.load_fhs**: Loads data from the FHS dataset.
```

--------------------------------

### Cache Management

Source: https://bio-learn.github.io/index.html

Utilities for managing data caching.

```APIDOC
## Cache Management

### Description
Utilities for managing data caching.

### Classes

#### `biolearn.cache.NoCache`

Represents no caching.

### Methods

(Specific methods not detailed in the provided text)

### Request Example

(Not available in the provided text)

### Response

(Not available in the provided text)

#### `biolearn.cache.LocalFolderCache`

Represents caching to a local folder.

### Methods

(Specific methods not detailed in the provided text)

### Request Example

(Not available in the provided text)

### Response

(Not available in the provided text)
```

--------------------------------

### LocalFolderCache.store

Source: https://bio-learn.github.io/modules/generated/biolearn.cache.LocalFolderCache.html

Stores a value in the cache, associated with a key, category, and version. Objects exceeding the cache size limit are not stored. Stale files in the same category but different versions are removed before storing.

```APIDOC
## LocalFolderCache.store

### Description
Store a value in the cache under the given key/category/version.

### Method
store

### Parameters
#### Path Parameters
- **key** (str) - Required - Unique key identifying the cache entry.
- **value** (Any) - Required - Object to store.
- **category** (str) - Required - Cache category name.
- **version** (str) - Required - Version string for invalidation.

### Notes
If the serialized object exceeds the cache size limit, it will not be stored. Before storing, stale files in the same category but different version are removed.
```

--------------------------------

### Predict Age using the Lin Model

Source: https://bio-learn.github.io/_downloads/097efd573efd69c248c612b7d5ada8b7/simple_competiton_submission.ipynb

Utilizes the 'Lin' model from ModelGallery to predict age based on the loaded challenge data. This model was selected for its performance.

```python
from biolearn.model_gallery import ModelGallery

#Lin scored the best so lets use that
results = ModelGallery().get("Lin").predict(challenge_data)
results
```

--------------------------------

### Cache Management

Source: https://bio-learn.github.io/clocks.html

Classes for managing data caching.

```APIDOC
## Cache Management

### Description
Utilities for handling data caching to improve performance.

### Classes

#### `biolearn.cache.NoCache`
Disables caching.

#### `biolearn.cache.LocalFolderCache`
Implements caching using a local folder.
```

--------------------------------

### Model Gallery Access

Source: https://bio-learn.github.io/clocks.html

Class for accessing the model gallery.

```APIDOC
## Model Gallery

### Description
Provides access to a gallery of pre-trained biological models.

### Class

#### `biolearn.model_gallery.ModelGallery`
Interface for browsing and retrieving models from the gallery.
```

--------------------------------

### Cache Management

Source: https://bio-learn.github.io/data.html

Classes for managing data caching to improve performance.

```APIDOC
## Cache Management

### Description
Utilities for handling data caching.

### Classes

#### `biolearn.cache.NoCache`
Disables caching.

#### `biolearn.cache.LocalFolderCache`
Uses a local folder for caching.
```

--------------------------------

### Load NHANES Data

Source: https://bio-learn.github.io/data.html

Use `load_nhanes` to load data from the National Health and Nutrition Examination Survey. Ensure the data is available in the expected format.

```python
from biolearn.load import load_nhanes

data = load_nhanes()
```

--------------------------------

### Predict Age with Decorated Clock and Verify Results

Source: https://bio-learn.github.io/_downloads/3e6fece343f1c608886b6406c2c75b77/plot_custom_imputation_implementation.ipynb

Uses the decorated clock model to predict epigenetic age and verifies that the results are identical to performing direct imputation followed by prediction.

```python
# Predict epigenetic age using the decorated clock (with imputation)
decorated_clock_predictions = decorated_clock.predict(target_dataset)

# Verify that the results are the same as the direct imputation approach
same_results = all(decorated_clock_predictions == imputed_age_predictions)
print(same_results)
```

--------------------------------

### Model Gallery Access

Source: https://bio-learn.github.io/data.html

Class for accessing a gallery of pre-trained biological models.

```APIDOC
## Model Gallery

### Description
Provides access to a collection of biological models.

### Class

#### `biolearn.model_gallery.ModelGallery`
Represents a gallery of models.
```

--------------------------------

### Visualize correlation matrix across aging models

Source: https://bio-learn.github.io/_downloads/108af8868614eea84b27d2927b9893eb/plot_clock_model_relationships.ipynb

Generates a correlation matrix heatmap for a specified set of aging models using a loaded GEO dataset.

```python
# Load an appropriate GEO dataset for the models
data = DataLibrary().get("GSE120307").load()

# Create a list of ModelGallery objects to be analyzed.
modelnames = ["Horvathv1", "Hannum", "PhenoAge", "DunedinPACE", "Lin", "Zhang_10"]
models = [ModelGallery().get(names) for names in modelnames]

plot_clock_correlation_matrix(
    models=models,
    data=data,
)
```

--------------------------------

### Cache API

Source: https://bio-learn.github.io/modules/generated/biolearn.cache.LocalFolderCache.html

Reference for cache management utilities.

```APIDOC
## Cache API

### Description
Utilities for managing data caching to improve performance.

### Classes
- **biolearn.cache.NoCache**: A cache implementation that does not store any data.
- **biolearn.cache.LocalFolderCache**: A cache implementation that stores data in a local folder.
```

--------------------------------

### GeoData Initialization

Source: https://bio-learn.github.io/data.html

Initialize `GeoData` to handle data from GEO (Gene Expression Omnibus). This class is specifically designed for GEO datasets.

```python
from biolearn.data_library import GeoData

geo_data = GeoData(accession='GSE12345')
```

--------------------------------

### Load local data file

Source: https://bio-learn.github.io/auto_examples/00_omics_biomarkers/plot_local_data_load.html

Loads methylation data from a local directory using the GeoData class. Ensure files follow the [name]_metadata.csv and [name]_methylation_part0.csv naming convention.

```python
from biolearn.data_library import GeoData
from biolearn.util import get_test_data_file

#Files formatted as described in the standard https://bio-learn.github.io/methylation-standard.html will load correctly.
#Load will search for files names [name]_metadata.csv and [name]_methylation_part0.csv

file_path = get_test_data_file("")
data = GeoData.load_csv(file_path, "example")

data.dnam
```

--------------------------------

### Predict Biological Age with Hurdle InflammAge

Source: https://bio-learn.github.io/_downloads/08e83fe39adcdfae93db7cbbc90ed77d/hurdle_api_example.ipynb

Requires the HURDLE_API_KEY environment variable to be set. The model will prompt for user consent before transmitting data to Hurdle's servers.

```python
import os
import pandas as pd
from biolearn.model_gallery import ModelGallery
from biolearn.data_library import DataLibrary

print("Loading sample data...")
data = DataLibrary().get("BoAChallengeData").load()

# Use first 5 samples for quick testing
sample_data = data.dnam.iloc[:, :5]
print(f"Using {sample_data.shape[1]} samples with {sample_data.shape[0]} CpG sites")

gallery = ModelGallery()

# Check if API key is set
if not os.environ.get("HURDLE_API_KEY"):
    print("\nNo HURDLE_API_KEY environment variable found.")
    api_key = input("Enter your Hurdle API key: ").strip()
    if api_key:
        os.environ["HURDLE_API_KEY"] = api_key
    else:
        print("No API key provided. Exiting.")
        exit(1)

# Load the model
model = gallery.get("HurdleInflammAge")

print("\nMaking predictions...")
print("Note: You will be prompted to consent to sending data to Hurdle's servers.")

try:
    predictions = model.predict(sample_data)

    print("\nInflammAge Results:")
    print("-" * 40)
    for sample, age in predictions.items():
        print(f"{sample}: {age:.1f} years")

    # If metadata is available, show age acceleration
    if hasattr(data, "metadata") and "age" in data.metadata.columns:
        print("\nAge Acceleration (InflammAge - Chronological Age):")
        print("-" * 40)
        for sample in predictions.index:
            if sample in data.metadata.index:
                inflamm_age = predictions[sample]
                chrono_age = data.metadata.loc[sample, "age"]
                delta = inflamm_age - chrono_age
                print(f"{sample}: {delta:+.1f} years")

except ValueError as e:
    print(f"Error: {e}")
except Exception as e:
    print(f"API Error: {e}")
    print("\nTroubleshooting:")
    print("- Verify your API key is correct")
    print("- Check your internet connection")
    print("- Ensure you have access to https://api.hurdle.bio")
```

--------------------------------

### Create Submission DataFrame

Source: https://bio-learn.github.io/_downloads/9d6419c1b7da427cabe6da2ae524be9e/training_simple_model.ipynb

Generates a Pandas DataFrame containing the predicted ages for the challenge data, indexed by 'sampleId', suitable for submission. Requires pandas.

```python
import pandas as pd

predicted_age_df = pd.DataFrame({
    'predictedAge': challenge_results
}, index=challenge_data.dnam.columns)
predicted_age_df.index.name = 'sampleId'
predicted_age_df
```

--------------------------------

### Predict and verify decorated clock results

Source: https://bio-learn.github.io/auto_examples/00_omics_biomarkers/plot_custom_imputation_implementation.html

Execute predictions using the decorated clock and compare them against direct imputation results.

```python
# Predict epigenetic age using the decorated clock (with imputation)
decorated_clock_predictions = decorated_clock.predict(target_dataset)

# Verify that the results are the same as the direct imputation approach
same_results = all(decorated_clock_predictions == imputed_age_predictions)
print(same_results)
```

```text
True
```

--------------------------------

### GeoData Methods

Source: https://bio-learn.github.io/modules/generated/biolearn.data_library.GeoData.html

Documentation for the methods available in the GeoData class.

```APIDOC
## GeoData Methods

### copy()

- **Description**: Creates a deep copy of the GeoData instance.
- **Returns**: A new instance of GeoData with copies of the metadata and dnam DataFrames.
- **Return type**: GeoData
```

```APIDOC
### quality_report(sites=None)

- **Description**: Generates a quality control report for the genomic data, optionally filtered by specified methylation sites, and includes a detailed section reporting the missing percentage for each methylation site.
- **Parameters**:
  - **sites** (list, optional) - A list of methylation site identifiers to include in the report. If None, all sites are included.
- **Returns**: An object containing both detailed methylation data, a summary, and a detailed section for missing percentages per site.
- **Return type**: QualityReport
```

```APIDOC
### from_methylation_matrix(matrix)

- **Description**: Creates a GeoData instance from a methylation matrix which can be either a DataFrame directly or a path to a CSV file.
- **Parameters**:
  - **matrix** (Union[str, DataFrame]) - Methylation matrix as a DataFrame or the path to the CSV file containing the matrix.
- **Returns**: An instance of GeoData with the methylation data loaded and metadata initialized.
- **Return type**: GeoData
```

```APIDOC
### save_csv(folder_path, name)

- **Description**: Saves the GeoData instance to CSV files according to the DNA Methylation Array Data Standard V-2410.
- **Parameters**:
  - **folder_path** (str) - The directory where the files will be saved.
  - **name** (str) - The base name for the saved files.
- **Returns**: None
```

```APIDOC
### load_csv(folder_path, name, series_part='all', validate=True)

- **Description**: Loads a GeoData instance from CSV files according to the DNA Methylation Array Data Standard V-2410.
- **Parameters**:
  - **folder_path** (str) - The directory where the files are located.
  - **name** (str) - The base name for the files.
  - **series_part** (str or int) - “all” to load all methylation parts and concatenate; otherwise, an integer specifying the part number to load.
  - **validate** (bool) - Whether to validate metadata-omics consistency. Default is True.
- **Returns**: A GeoData instance with metadata, methylation data, RNA, and protein data loaded.
- **Return type**: GeoData
```

--------------------------------

### DataSource Class Reference

Source: https://bio-learn.github.io/modules/generated/biolearn.data_library.DataSource.html

Overview of the DataSource class methods and attributes for data management.

```APIDOC
## DataSource Class

### Description
The `DataSource` class is the primary interface for interacting with epigenetic data libraries. It handles the loading of datasets and manages cache categories and versions.

### Attributes
- **REQUIRED_FIELDS** (list) - List of fields that must be present in the data source.
- **CACHE_CATEGORY** (string) - The category identifier used for caching data.
- **CACHE_VERSION** (string) - The version identifier for the cached data.

### Methods
- **__init__** - Initializes a new instance of the DataSource.
- **load** - Loads the data from the source into the application environment.
```

--------------------------------

### ModelGallery.search

Source: https://bio-learn.github.io/modules/generated/biolearn.model_gallery.ModelGallery.html

Search models based on species and tissue criteria.

```APIDOC
## ModelGallery.search

### Description
Searches for models based on species and tissue criteria.

### Method
GET

### Endpoint
/models/search

### Parameters
#### Query Parameters
- **species** (str) - Optional - The species to filter models by.
- **tissue** (str) - Optional - The tissue to filter models by.

### Request Example
```json
{
  "species": "human",
  "tissue": "liver"
}
```

### Response
#### Success Response (200)
- **models** (dict) - A dictionary of models that match the specified criteria.

### Response Example
```json
{
  "model_name_1": { ... },
  "model_name_2": { ... }
}
```
```

--------------------------------

### Visualize model predictions against chronological age

Source: https://bio-learn.github.io/_downloads/108af8868614eea84b27d2927b9893eb/plot_clock_model_relationships.ipynb

Plots predicted ages from selected models against actual chronological age.

```python
# Use appropriate clocks/models
modelnames = ["Horvathv1", "Hannum", "PhenoAge", "Lin"]
age_prediction_models = [ModelGallery().get(name) for name in modelnames]

plot_age_prediction(
    models=age_prediction_models,
    data=data,
)
```

--------------------------------

### biolearn.cache : Cache

Source: https://bio-learn.github.io/modules/generated/biolearn.model_gallery.ModelGallery.html

Reference for caching mechanisms.

```APIDOC
## biolearn.cache.NoCache

### Description
Represents a cache that does not store any data.

### Parameters
This class does not have explicit parameters listed in the provided text.

## biolearn.cache.LocalFolderCache

### Description
Implements a cache that stores data in a local folder.

### Parameters
This class does not have explicit parameters listed in the provided text.
```

--------------------------------

### biolearn.cache.LocalFolderCache

Source: https://bio-learn.github.io/modules/cache.html

An LRU-style cache that stores pickled objects in a specified local directory.

```APIDOC
## Class: biolearn.cache.LocalFolderCache

### Description
LRU-style cache that stores pickled objects in a local directory.

### Constructor
`LocalFolderCache(path, max_size_gb)`

### Parameters
- **path** (string) - Required - The local directory path where cached objects are stored.
- **max_size_gb** (float) - Required - The maximum size of the cache in gigabytes.
```

--------------------------------

### Visualize methylation by age and sex

Source: https://bio-learn.github.io/_downloads/a78e2414f22fb790bca7559350d8cb52/plot_dnam_relationships.ipynb

Compute age group statistics and generate a violin plot for a specific CpG site.

```python
# Initialize an appropriate CpG site
cpg_site = "cg00000029"

# Compute statistics by age group for the specified CpG site
stats_df, combined_data1 = compute_age_group_stats(datasets, cpg_site, age_bins=5)

# Plot DNA methylation levels by age group and sex across datasets in a violin plot
# Note: In the datasets, '1' represents Female and '2' represents Male.
plot_methylation_by_age_sex(combined_data1, cpg_site)
```

--------------------------------

### Visualize methylation vs age with regression

Source: https://bio-learn.github.io/_downloads/a78e2414f22fb790bca7559350d8cb52/plot_dnam_relationships.ipynb

Compute regression statistics and plot DNA methylation against age using linear and polynomial regression lines.

```python
# Compute regression statistics to plot methylation vs. age
combined_data2, stats_df = compute_methylation_stats(datasets, cpg_site)

# Plot DNA methylation against age across datasets with linear and polynomial regression lines
plot_methylation_vs_age(combined_data2, stats_df, cpg_site)
```

--------------------------------

### View Combined Metadata

Source: https://bio-learn.github.io/_downloads/e2c40c72c94a18149a91c1d47ee4de74/plot_challenge_data.ipynb

Displays the combined metadata associated with the challenge data. This includes all metadata, even if data points do not overlap between different data types.

```python
challenge_data.metadata
```