write.narrowpeak.binding / columns

[orchid00]

Hi,

I'm using version spp_1.14.
I'm trying to understand the function write.narrowpeak.binding.
The code is as follows:
bp <- add.broad.peak.regions(chip.data, input.data, bp, window.size = 1000, z.thr = 3)

Write out determined binding peaks using narrowPeak format

write.narrowpeak.binding(bp, paste0(outfile,".narrowPeak"))

Does the function write.narrowpeak.binding filters some of the broad peaks? as called from the previous function add.broad.peak.regions?

Another question is could you please give me the header names of the output file for narrowPeak
I have 10 columns
str(narrowpeaks)
'data.frame': 123317 obs. of 10 variables:
$ V1 : Factor w/ 89 levels "chr1","chr1_gl000191_random",..: 66 48 66 66 66 48 66 66 66 66 ...
$ V2 : int 7 32797 7 7 7 32797 7 7 7 7 ...
$ V3 : int 15981 127412 15981 15981 15981 127412 15981 15981 15981 15981 ...
$ V4 : Factor w/ 1 level ".": 1 1 1 1 1 1 1 1 1 1 ...
$ V5 : int 0 0 0 0 0 0 0 0 0 0 ...
$ V6 : Factor w/ 1 level ".": 1 1 1 1 1 1 1 1 1 1 ...
$ V7 : num 2690 2072 2033 2004 2003 ...
$ V8 : int -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 ...
$ V9 : num 4.88 4.88 4.88 4.88 4.88 ...
$ V10: int 14338 50301 5946 13411 1113 55228 6570 7004 14754 136 ...

in comparison to 9 columns in the broadPeaks (using: write.broadpeak.info(broad.clusters, paste0(outfile,".broadPeak"))

str(broadpeaks)
'data.frame': 45304 obs. of 9 variables:
$ V1: Factor w/ 80 levels "chr1","chr1_gl000191_random",..: 18 37 26 79 18 25 22 34 35 33 ...
$ V2: int 48109886 3054837 48342560 1709365 289562 184278556 46705634 4101231 1201968 2221121 ...
$ V3: int 48114823 3058716 48345231 1712527 292422 184281576 46709671 4104549 1204422 2225464 ...
$ V4: Factor w/ 1 level ".": 1 1 1 1 1 1 1 1 1 1 ...
$ V5: int 0 0 0 0 0 0 0 0 0 0 ...
$ V6: Factor w/ 1 level ".": 1 1 1 1 1 1 1 1 1 1 ...
$ V7: num 8.49 8.49 8.32 8.31 8.29 ...
$ V8: int -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 ...
$ V9: int -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 ...

I'm interested to know which columns correspond to start and end position, I assumed it will be 2 and 3 for both cases, but now I think it is not the case.

[pkharchenko]

Hi there,
Yes, spp looks for broader peaks to write out narrowPeak format. First, here’s the description of the format (columns 1-10 in the output data.frame correspond to the fields 1-10): http://genome.ucsc.edu/FAQ/FAQformat.html#format12 http://genome.ucsc.edu/FAQ/FAQformat.html#format12

spp considers peak a singular local maximum. It may have a restriction that it has to be the highest peak within x basepairs (see min.dist parameter in find.binding.positions). This corresponds to column 10 (peak) of the narrowPeak format.

However, narrowPeak also reports chromStart/chromEnd (columns 2,3) that describe the extent of the broader region of enrichment around the peak. That’s what this code is trying to do.

Hope that helps,
-peter.

On Aug 25, 2016, at 12:19 AM, Paula Andrea notifications@github.com wrote:

Hi,

I'm using version spp_1.14.
I'm trying to understand the function write.narrowpeak.binding.
The code is as follows:
bp <- add.broad.peak.regions(chip.data, input.data, bp, window.size = 1000, z.thr = 3)

Write out determined binding peaks using narrowPeak format

write.narrowpeak.binding(bp, paste0(outfile,".narrowPeak"))

Does the function write.narrowpeak.binding filters some of the broad peaks? as called from the previous function add.broad.peak.regions?

Another question is could you please give me the header names of the output file for narrowPeak
I have 10 columns
str(narrowpeaks)
'data.frame': 123317 obs. of 10 variables:
$ V1 : Factor w/ 89 levels "chr1","chr1_gl000191_random",..: 66 48 66 66 66 48 66 66 66 66 ...
$ V2 : int 7 32797 7 7 7 32797 7 7 7 7 ...
$ V3 : int 15981 127412 15981 15981 15981 127412 15981 15981 15981 15981 ...
$ V4 : Factor w/ 1 level ".": 1 1 1 1 1 1 1 1 1 1 ...
$ V5 : int 0 0 0 0 0 0 0 0 0 0 ...
$ V6 : Factor w/ 1 level ".": 1 1 1 1 1 1 1 1 1 1 ...
$ V7 : num 2690 2072 2033 2004 2003 ...
$ V8 : int -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 ...
$ V9 : num 4.88 4.88 4.88 4.88 4.88 ...
$ V10: int 14338 50301 5946 13411 1113 55228 6570 7004 14754 136 ...

in comparison to 9 columns in the broadPeaks (using: write.broadpeak.info(broad.clusters, paste0(outfile,".broadPeak"))

str(broadpeaks)
'data.frame': 45304 obs. of 9 variables:
$ V1: Factor w/ 80 levels "chr1","chr1_gl000191_random",..: 18 37 26 79 18 25 22 34 35 33 ...
$ V2: int 48109886 3054837 48342560 1709365 289562 184278556 46705634 4101231 1201968 2221121 ...
$ V3: int 48114823 3058716 48345231 1712527 292422 184281576 46709671 4104549 1204422 2225464 ...
$ V4: Factor w/ 1 level ".": 1 1 1 1 1 1 1 1 1 1 ...
$ V5: int 0 0 0 0 0 0 0 0 0 0 ...
$ V6: Factor w/ 1 level ".": 1 1 1 1 1 1 1 1 1 1 ...
$ V7: num 8.49 8.49 8.32 8.31 8.29 ...
$ V8: int -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 ...
$ V9: int -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 ...

I'm interested to know which columns correspond to start and end position, I assumed it will be 2 and 3 for both cases, but now I think it is not the case.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub #6, or mute the thread https://github.com/notifications/unsubscribe-auth/ALT78ro-3TNsDyqvCV8dyVn17pfslYFSks5qjRfigaJpZM4Jsrk2.

[orchid00]

Thank you Peter,
I appreciate your quick reply. I understand the narrow peaks much better now. Also, thanks for the links, that also confirms columns 2 and 3 are star and end which is what I need.
kind regards,

Paula