A new version of pyBigWig. Compared with the previous version, the new version improves the way handling null values in bigwig files.
You can install this extension directly from github with:
pip install git+https://github.com/1944498970/pyBigWig.git
The usage is same with the previous version, you can read the readme.md of previous version.
If our bedgraph file is like:
track type=bedGraph name="BedGraph Format" description="BedGraph format" visibility=full color=200,100,0 altColor=0,100,200 priority=20
chr1 4 8 -1.0
chr1 16 20 -0.75
chr1 25 30 -0.50
We turn this file into bigwig format named test.bigwig.Then we use pyBigWig to open it:
import pyBigWig
bw=pyBigWig.open('test.bigwig')
Using some commands of pyBigWig to handle it, the final results of two versions will be list as follows.
| command | pyBigWig-0.3.19 | pyBigWig-0.3.18 |
|---|---|---|
| bw.values('chr1' , 1, 4) | [0.0, 0.0, 0.0] | [nan, nan, nan] |
| bw.stats('chr1' , 1, 4) | [0.0] | [None] |
| bw.stats('chr1' , 4, 8) | [-1.0] | [-1.0] |
| bw.stats('chr1' , 4, 16) | [-0.3333333333333333] | [-1.0] |
| bw.stats('chr1' , 1, 4,nBins=2) | [0.0, 0.0] | [None, None] |
| bw.stats('chr1' , 4, 8,nBins=2) | [-1.0, -1.0] | [-1.0, -1.0] |
| bw.stats('chr1' , 4, 16,nBins=4)) | [-1.0, -0.3333333333333333, 0.0, 0.0] | [-1.0, -1.0, None, None] |
According to the list we can find the previous version can't handle the site without value, so when we get its value it will print nan,and when we calculate the mean of the region of [4,16] the previous version will ignore sites without values so it will print -1.0 which is not same as the true mean.