Biomedical Engineering Reference
In-Depth Information
In the example above, we used a total of 200 bins (option '-b 200'),
and a smoothing parameter '-m 10', which sums the results in each series
of consecutive 10 bins.
8.5.3 Creating window-based read densities
In ChIP-seq studies, researchers are interested in visualizing the densities
of their ChIP-seq reads at a genome-wide scale so that they can understand
the behavior of the studied protein in genes of interest. GenomicTools can
be used to create window-based read densities to be displayed as a wiggle
track in the UCSC Genome Browser (Figure 8.7). First, the user needs to
decide on the window parameters: (1) the size of the window; (2) the
distance between consecutive windows; and (3) minimum number of read
allowed in each window. The last two parameters establish a tradeoff
between resolution and output fi le size. Here are some typical values:
$ set win_size = 500
$ set win_size = 500
# must be a multiple of win_dist
# must be a multiple of win_dist
$ set win_dist = 25
$ set min_reads = 20
$ set win_dist = 25
$ set min_reads = 20
Then, the user needs to create a fi le describing the chromosomal bounds
in REG or BED format:
$ head genome.bed
chr1 0 197195432
...
$ head genome.bed
chr1 0 197195432
. . .
The genomic_scans tool can be used to compute the counts of reads
stored in 'chipseq.bed' in sliding windows across the genome. Finally, the
center of each window is computed and the '-wig' operation of genomic_
regions converts to the wiggle format for display in the UCSC Genome
Browser (Figure 8.7):
$ head chipseq.bed
chr1 3001228 3001229
chr1 3001228 3001229
chr1 3001438 3001439
. . .
$ head chipseq.bed
chr1 3001228 3001229
chr1 3001228 3001229
chr1 3001438 3001439
. . .
$ cat chipseq.bed | genomic_scans counts -v -min $min_reads -w $win_size
-d $win_dist -g genome.bed | genomic_regions center | genomic_regions
wig -t 'densities' -s $win_dist -c '0,0,150' > densities.wig
$ cat chipseq.bed | genomic_scans counts -v -min $min_reads -w $win_size
-d $win_dist -g genome.bed | genomic_regions center | genomic_regions
wig -t 'densities' -s $win_dist -c '0,0,150' > densities.wig
$ head densities.wig
$ head densities.wig
Search WWH ::
Custom Search