DADApy

DADApy analyzes high-dimensional data manifolds to estimate intrinsic dimensionality, compute probability density, perform density-based clustering, and compare distance metrics for characterization of complex datasets in bioinformatics, genomics, and systems biology.

Key Features:

• Intrinsic Dimension Estimation: Provides methods to estimate the intrinsic dimensionality of data manifolds, identifying the minimal number of variables required to represent the dataset.
• Probability Density Estimation: Estimates probability density functions on data manifolds to identify regions of varying data concentration.
• Density-Based Clustering: Performs clustering based on local data density to group points while accounting for density variations.
• Distance Metric Comparison: Enables comparison of different distance metrics to select metrics that align with the data's inherent manifold structure.

Scientific Applications:

• Bioinformatics: Characterizes high-dimensional omics and other bioinformatics datasets through manifold-aware density and dimensionality analyses.
• Genomics: Estimates intrinsic dimensions and clusters genetic datasets to reveal patterns related to diseases or traits.
• Systems Biology: Analyzes complex multivariate systems to uncover underlying manifold structures and heterogeneities.

Methodology:

DADApy employs distance-based methods and advanced statistical and computational techniques and supports analysis of both synthetic and real-world datasets.