wtsi-hgi/lofi

Name: lofi

Owner: Wellcome Trust Sanger Institute - Human Genetics Informatics

Description: Sequence alignment data downsampling tool

Created: 2018-03-02 10:55:01.0

Updated: 2018-03-05 16:03:02.0

Pushed: 2018-03-05 16:02:48.0

Homepage:

Size: 23

Language: Shell

GitHub Committers

User	Most Recent Commit	# Commits

Other Committers

User	Email	Most Recent Commit	# Commits

README

Lo-Fi

Downsample sequence alignment data (SAM, BAM or CRAM files) given specific constraints.

Requirements

• GNU Awk
• Samtools 1.?

Usage

lofi [--coverage COVERAGE] [--species SPECIES]
     [--read-groups RG_RANGE] [--rg-coverage RG_COVERAGE] [--keep-rg RG]
     [--strategy STRATEGY] [--tolerance TOLERANCE] [--seed SEED]
     [--prefix PREFIX] [--no-recombine]
     INPUT

At least one constraint must be specified.

Coverage Constraints

--coverage COVERAGE        The overall coverage of the output
--rg-coverage RG_COVERAGE  The minimum coverage in output read groups

COVERAGE and RG_COVERAGE are specified as the number of base pairs, optionally suffixed with k, M or G. If the --species option is used, an x suffix may also be used, indicating a coverage relative to the species genome size (under the presumption that your input is whole genome sequence data).

Supported SPECIES values:

• human
• mouse
• zebrafish

Read Group Constraints

--read-groups RG_RANGE     Read groups in the output
--keep-rg RG               Do not drop the read group with the given ID;
                           this can be specified multiple times

The RG_RANGE takes the format MIN-, -MAX, MIN-MAX or EXACT.

Downsampling Options

--strategy STRATEGY        Downsampling strategy [default: DROP_RG]
--tolerance TOLERANCE      Allowed percentage difference in output
                           coverage from target [default: 10]
--seed SEED                Random seed [default: 0]

Supported STRATEGY values:

• DROP_RG: Prefer dropping read groups over downsampling reads

Output Options

--prefix PREFIX            Output path prefix
--no-recombine             Don't recombine individual read groups

The output file(s) will be named [PREFIX]INPUT[.READ_GROUP].downsampled, where READ_GROUP will be the read group ID if --no-recombine is specified. The PREFIX, if specified, is relative to the INPUT's location, or an absolute path.