The goal of metaseq is to tie together lots of existing software into a framework for exploring genomic data. It focuses on flexibility and interactive exploration of data.

metaseq is sort of like a glue that brings together lot of great software commonly used for genomic data analysis. It gives you the tools you need to do analyses that are difficult to do in any other way that I know of. It ties together other packages like:

• Pysam¹ for access to BAM and SAM files
• bx-python² for access to bigWig and bigBed files
• pybedtools³, which integrates BEDTools into Python, for fast "genome algebra"
• matplotlib⁴ for rich, interactive, highly customizable plotting
• gffutils⁵ for hierarchical access to gene models
• Cython⁶ for speed
• scikits.learn⁷ for fast clustering
• NumPy⁸ for large, fast arrays

metaseq also:

• Supplies a fast "rebinning" routine that is compiled in C for speed
• Accepts bigBed, bigWig, BAM and SAM files as input for random access of continuous data (and with Tabix, accepts BED, GFF, GTF, VCF as well)
• Support for tunable (i.e., tradeoffs between CPU, RAM, and I/O) parallel processing of data accessed from the file formats mentioned above
• Provides a framework for "mini-browsers", zoomable and pannable Python-only figures that show genomic signal and gene models and are spawned by clicking on features of interest
• Integrates RNA-seq and ChIP-seq data by using gffutils to convert between gene IDs and coordinates as well as handle full gene models and plotting.