hammerlab/bai-indexer

Name: bai-indexer

Owner: Hammer Lab

Description: Build an index for your BAM Index (BAI)

Created: 2014-10-24 17:17:36.0

Updated: 2016-10-13 23:00:50.0

Pushed: 2015-04-14 17:35:57.0

Homepage: null

Size: 212

Language: Python

GitHub Committers

User	Most Recent Commit	# Commits

Other Committers

User	Email	Most Recent Commit	# Commits

README

bai-indexer

Build an index for your BAM Index (BAI).

Background

BAM is a common file format for storing aligned reads from a gene sequencing machine. These files can get enormous (100+ GB), so it's helpful to have an index to support fast lookup.

Samtools defines a file format for a BAM index and provides a simple command for generating one:

ools index file.bam file.bam.bai

Unfortunately, these BAM Index (BAI) files can also grow very large, often to 10 MB or more. When using a genome browser like IGV or BioDalliance, loading a large BAI file over a slow network is the unavoidable first step in displaying alignment tracks.

bai-indexer solves this problem by building an index of your BAM Index. This is a small JSON file which maps reference ID (i.e. chromosome number) to a byte range within the BAI file. By loading the BAM index, a viewer can load only the small subset of the BAM index that it actually needs.

Usage

pip install bai-indexer

bai-indexer path/to/file.bam.bai > path/to/file.bam.bai.json

Format

The JSON index index looks like this:

hunks": [
[8, 716520],
[716520, 1463832],
[1463832, 2070072],
...

inBlockIndex": 1234

The first chunk ([8, 716520]) specifies the byte range in the BAI file which describes the first ref (most likely chr1 for a human genome). This is a half-open [start, stop) interval.

The minBlockIndex field specifies the position of the first block in the BAM file. Everything before this position is headers.

Development

After setting up a virtualenv, you can get going by running:

install -r requirements.txt
tests