Scroll to navigation

RABEMA_BUILD_GOLD_STANDARD(1) RABEMA_BUILD_GOLD_STANDARD(1)

NAME

rabema_build_gold_standard - RABEMA Gold Standard Builder

SYNOPSIS

rabema_build_gold_standard [OPTIONS] --out-gsi OUT.gsi --reference REF.fa --in-bam PERFECT.{sam,bam}

DESCRIPTION

This program allows one to build a RABEMA gold standard. The input is a reference FASTA file and a perfect SAM/BAM map (e.g. created using RazerS 3 in full-sensitivity mode).

The input SAM/BAM file must be sorted by coordinate. The program will create a FASTA index file REF.fa.fai for fast random access to the reference.

OPTIONS

-h, --help
Display the help message.
--version
Display version information.
-v, --verbose
Enable verbose output.
-vv, --very-verbose
Enable even more verbose output.

Input / Output:

-o, --out-gsi OUTPUT_FILE
Path to write the resulting GSI file to. Valid filetype is: .gsi[.*], where * is any of the following extensions: gz for transparent (de)compression.
-r, --reference INPUT_FILE
Path to load reference FASTA from. Valid filetypes are: .sam[.*], .raw[.*], .gbk[.*], .frn[.*], .fq[.*], .fna[.*], .ffn[.*], .fastq[.*], .fasta[.*], .faa[.*], .fa[.*], .embl[.*], and .bam, where * is any of the following extensions: gz, bz2, and bgzf for transparent (de)compression.
-b, --in-bam INPUT_FILE
Path to load the "perfect" SAM/BAM file from. Valid filetypes are: .sam[.*] and .bam, where * is any of the following extensions: gz, bz2, and bgzf for transparent (de)compression.

Gold Standard Parameters:

--oracle-mode
Enable oracle mode. This is used for simulated data when the input SAM/BAM file gives exactly one position that is considered as the true sample position.
--match-N
When set, N matches all characters without penalty.
--distance-metric STRING
Set distance metric. Valid values: hamming, edit. Default: edit. One of hamming and edit. Default: edit.
-e, --max-error INTEGER
Maximal error rate to build gold standard for in percent. This parameter is an integer and relative to the read length. In case of oracle mode, the error rate for the read at the sampling position is used and RATE is used as a cutoff threshold. Default: 0.

RETURN VALUES

A return value of 0 indicates success, any other value indicates an error.

EXAMPLES

rabema_build_gold_standard -e 4 -o OUT.gsi -s IN.sam -r REF.fa
Build gold standard from a SAM file IN.sam with all mapping locations and a FASTA reference REF.fa to GSI file OUT.gsi with a maximal error rate of 4.
rabema_build_gold_standard --distance-metric edit -e 4 -o OUT.gsi -b IN.bam -r REF.fa
Same as above, but using Hamming instead of edit distance and BAM as the input.
rabema_build_gold_standard --oracle-mode -o OUT.gsi -s IN.sam -r REF.fa
Build gold standard from a SAM file IN.sam with the original sample position, e.g. as exported by read simulator Mason.

MEMORY REQUIREMENTS

From version 1.1, great care has been taken to keep the memory requirements as low as possible. There memory required is two times the size of the largest chromosome plus some constant memory for each match.

For example, the memory usage for 100bp human genome reads at 5% error rate was 1.7GB. Of this, roughly 400GB came from the chromosome and 1.3GB from the matches.

REFERENCES

M. Holtgrewe, A.-K. Emde, D. Weese and K. Reinert. A Novel And Well-Defined Benchmarking Method For Second Generation Read Mapping, BMC Bioinformatics 2011, 12:210.
http://www.seqan.de/rabema
RABEMA Homepage
http://www.seqan.de/mason
Mason Homepage
rabema_build_gold_standard 1.2.9 [tarball]