Manages annotated data matrices and single-cell genomics datasets using the AnnData framework in Python.
This skill provides specialized guidance for working with AnnData objects, the foundational data structure for single-cell genomics and large-scale biological measurements. It assists users in creating, manipulating, and storing complex datasets with multi-dimensional annotations (obs, var, obsm, layers), ensuring full compatibility with the scverse ecosystem tools like Scanpy and scvi-tools. Whether you are performing quality control, subsetting massive datasets in memory-efficient backed mode, or concatenating multiple experimental batches, this skill offers the implementation patterns and performance-optimized practices required for robust biological data analysis.
主要功能
01Efficient data manipulation, subsetting, and filtering techniques
02Advanced dataset concatenation and batch integration strategies
03High-performance I/O for h5ad, zarr, loom, and 10X formats
04Memory-efficient practices including sparse matrices and backed mode
05Comprehensive AnnData object creation and metadata management
060 GitHub stars
使用场景
01Integrating multiple experimental batches into unified annotated datasets
02Single-cell RNA-seq preprocessing, normalization, and quality control
03Analyzing massive genomics datasets that exceed available system memory
