Source code for PrimerNuc online tool available at http://www.primer-suite.com/primernuc/
This code requires Python 3 to run.
Tested on Spyder (Python 3.8) running on Windows 7 Enterprise operating system
Tested using Python 3.4.3 on Linux server.
Description:
PrimerNuc is a search tool that enables users to locate all predicted nucleosome protection peaks within an input chromosomal region or the two nearest protection peaks flanking a specified nucleotide position. Along with peak positions, this software outputs the adjusted WPS (WPS divided by the mean WPS of the sample pool), the adjusted sequencing depth (coverage divided by the mean coverage of the sample pool), and distances (bp) of the output peak positions to the two input nucleotide positions.
Instructions:
Insert chromosomal regions (GRCh37/hg19 assembly) to search for nucleosome protection peaks into the input.bed file
Run the PrimerNucleosome.py script
Open the output.txt file for results
Results table is formatted with the following columns:
Chromosome #chromosome of search region from input.bed
Start position (hg19) #first base of searched region from input.bed
End position (hg19) #last base of search region from input.bed
Region descriptor #fourth column of input.bed
Sample #name of sample being search (i.e., CH01, BRA01, BRE01, COL01, BBC01)
Nucleosome peak position (hg19) #Protection peak position
Distance from start (bp) #distance of protection peak from first base of search region
Distance from end (bp) #distance of protection peak from first base of search region
Within or Flanking? #whether the peak inside the searhed region or is one of the two peaks flanking the searched region
WPS (mean = 1) #Adjusted windowed protection score (WPS divided by the mean WPS of the sample (e.g. 0.5 = half mean WPS, 2 = twice mean)
Read depth (mean = 1) #mean coverage of 180 bp around nucleosome protection peak divided by mean coverage of all bases in the sample (e.g. 0.06 = 6% of the mean coverage, 4.5 = 450% of the mean coverage)
Example input:
chr1 10000000 10000700 region
chr1 1000000 1000000 single base
Example output:
chr1 10000000 10000700 region BRA01 9999792 -208 -908 5' flanking 'n/a' 0.35
chr1 10000000 10000700 region BRA01 10000007 7 -693 within 'n/a' 0.38
chr1 10000000 10000700 region BRA01 10000222 222 -478 within 'n/a' 0.26
chr1 10000000 10000700 region BRA01 10000439 439 -261 within 0.96 0.24
chr1 10000000 10000700 region BRA01 10000658 658 -42 within 0.74 0.22
chr1 10000000 10000700 region BRA01 10000750 750 50 3' flanking 0.64 0.21
chr1 1000000 1000000 single base BRA01 999850 -150 -150 5' flanking 1.28 0.18
chr1 1000000 1000000 single base BRA01 1000036 36 36 3' flanking 'n/a' 0.16
Note: "n/a" in the WPS column means no peak exists but greather than 370 bp (twice the mode interpeak distance from the CH01 and CA01 comprehnsive nucleosome maps of Snyder et al. (2016)) seperates the two nearest nucleosome protection peaks in this region. Additional points are placed in these regions for users to check if the lack of peak is due either to poor sequencing coverage or broader nucleosome protection.
Samples include:
CH01 #nucleosome protection peaks from Snyder et al. (2016)
BRA01 #Brain cancer sample pool
BRE01 #Breast cancer sample pool
COL01 #Colorectal cancer sample pool
BBC01 #Pool of BRA01, BRE01 and COL01
To alter the sample being searched, edit the PrimerNucleosome.py script file at line 222:
results = Nucleosome("input.bed", "output.txt", "BRA01") # input file, output file, sample name
References:
Snyder, M. W., Kircher, M., Hill, A. J., Daza, R. M., & Shendure, J. (2016). Cell-free DNA comprises an in vivo nucleosome footprint that informs its tissues-of-origin. Cell, 164(1-2), 57-68.