This paper describes the Bioconductor infrastructure for representing and computing on annotated genomic ranges, integrating genomic data with R's statistical computing features. The infrastructure includes three core packages: IRanges, GenomicRanges, and GenomicFeatures. These packages provide scalable data structures for genomic ranges, with special support for transcript structures, read alignments, and coverage vectors. They include efficient algorithms for overlap detection, coverage calculation, and other range operations, directly supporting over 80 other Bioconductor packages, including those for sequence analysis, differential expression, and visualization. The infrastructure allows for the storage and manipulation of genomic ranges with metadata, such as gene identifiers and peak heights. It supports hierarchical ranges and integrates with other R packages through in-memory data structures, while also supporting interaction with external tools. The IRanges package provides fundamental range data structures, while GenomicRanges adds biological semantics, including strand and sequence name. GenomicFeatures enables access to and manipulation of gene models and annotations. The infrastructure supports a wide range of genomic data analyses, including the manipulation of gene model annotations, analysis of experimental data, and integration of genomic data with statistical tools. It provides efficient range-based algorithms and data structures, enabling tasks such as overlap detection, coverage calculation, and annotation of ChIP-seq and RNA-seq data. The infrastructure is designed to be efficient, scalable, and interoperable, with a focus on robustness and maintainability. The paper also describes the design and implementation of the infrastructure, including the use of classes to represent data structures and the use of generic functions to provide specialized behavior. It discusses the use of range-based operations for genomic data, including the use of the GRanges class for genomic ranges and the GRangesList class for grouping ranges. The infrastructure supports access to gene models and annotations, including the use of the GenomicFeatures package to distill multiple data sources into a single database schema. The paper also describes the use of the infrastructure for analyzing read alignments, coverage, and variant calling. It includes examples of using the infrastructure for analyzing ChIP-seq data, RNA-seq data, and other genomic data. The infrastructure is also used for summarizing genomic data, such as in the SummarizedExperiment class, which holds summaries of genomic data along with annotations. The paper concludes with a discussion of the software based on the infrastructure, including a growing ecosystem of packages that depend on the infrastructure. It also discusses the availability and future directions of the infrastructure, including the need for better visualization of genomic ranges and more efficient algorithms and data structures for handling increasingly complex and heterogeneous data.This paper describes the Bioconductor infrastructure for representing and computing on annotated genomic ranges, integrating genomic data with R's statistical computing features. The infrastructure includes three core packages: IRanges, GenomicRanges, and GenomicFeatures. These packages provide scalable data structures for genomic ranges, with special support for transcript structures, read alignments, and coverage vectors. They include efficient algorithms for overlap detection, coverage calculation, and other range operations, directly supporting over 80 other Bioconductor packages, including those for sequence analysis, differential expression, and visualization. The infrastructure allows for the storage and manipulation of genomic ranges with metadata, such as gene identifiers and peak heights. It supports hierarchical ranges and integrates with other R packages through in-memory data structures, while also supporting interaction with external tools. The IRanges package provides fundamental range data structures, while GenomicRanges adds biological semantics, including strand and sequence name. GenomicFeatures enables access to and manipulation of gene models and annotations. The infrastructure supports a wide range of genomic data analyses, including the manipulation of gene model annotations, analysis of experimental data, and integration of genomic data with statistical tools. It provides efficient range-based algorithms and data structures, enabling tasks such as overlap detection, coverage calculation, and annotation of ChIP-seq and RNA-seq data. The infrastructure is designed to be efficient, scalable, and interoperable, with a focus on robustness and maintainability. The paper also describes the design and implementation of the infrastructure, including the use of classes to represent data structures and the use of generic functions to provide specialized behavior. It discusses the use of range-based operations for genomic data, including the use of the GRanges class for genomic ranges and the GRangesList class for grouping ranges. The infrastructure supports access to gene models and annotations, including the use of the GenomicFeatures package to distill multiple data sources into a single database schema. The paper also describes the use of the infrastructure for analyzing read alignments, coverage, and variant calling. It includes examples of using the infrastructure for analyzing ChIP-seq data, RNA-seq data, and other genomic data. The infrastructure is also used for summarizing genomic data, such as in the SummarizedExperiment class, which holds summaries of genomic data along with annotations. The paper concludes with a discussion of the software based on the infrastructure, including a growing ecosystem of packages that depend on the infrastructure. It also discusses the availability and future directions of the infrastructure, including the need for better visualization of genomic ranges and more efficient algorithms and data structures for handling increasingly complex and heterogeneous data.