polars-bio

To find overlapping genomic intervals in large BED files, you can use lazy DataFrame execution to process datasets out-of-core. This approach computes overlaps, nearest features, and coverage efficiently without loading entire datasets into memory.

Yes, you can read and scan bioinformatics formats including BAM, CRAM, VCF, and BED directly from local or cloud paths into Polars DataFrames. This enables scalable file I/O for genomic interval analytics workflows.

Polars supports streaming execution for bioinformatics data when using a DataFusion-backed lazy execution model. This allows you to process large genomic datasets out-of-core and return LazyFrame results suitable for .collect().

Running SQL queries over genomic interval tables is best handled by registering interval DataFrames with DataFusion SQL. This integrates overlap, merge, and coverage computations directly into your SQL-based data analysis workflows.

Merge and cluster operations on genomic intervals are performed by applying interval arithmetic directly within Polars workflows. You can configure interval column conventions to compute merges, clusters, complements, and subtractions efficiently.

Interval arithmetic for high-throughput genomics requires a Polars-DataFusion based environment and returns LazyFrame results, meaning you must call .collect() to materialize outputs. Proper configuration of interval column conventions is necessary to ensure accurate overlap and coverage computations.