What problem does it solve?
This Skill solves slow, memory-heavy genomic interval computations by providing high-performance overlap, nearest, merge, coverage, complement, and subtract operations directly on Polars DataFrames.
Core Features & Use Cases
- Genomic interval arithmetic on Polars with output as LazyFrames by default, covering overlap, nearest, count overlaps, merge, cluster, coverage, complement, and subtract for BED-style coordinates.
- Bioinformatics file I/O for common genomics formats (BED, VCF, BAM/CRAM, GFF/GTF, FASTA/FASTQ, SAM, Hi-C pairs) with eager reads and streaming scans for large files.
- DataFusion SQL integration to register files/data as SQL tables and query them with standard SQL before running interval operations, plus pileup depth calculation from BAM/CRAM.
Quick Start
Use the polars-bio skill to compute streaming overlaps between two BED files named 'regions.bed' and 'genes.bed' and return the result as a Polars DataFrame.