Columns with all zero entries and negative chromosomal positions in bed files on the website

[soerenmueller]

Hi guys,

I hope you can help me with the following issue: I have some questions in regard to the data processing. i tried to get PSYCHIC to run of the data from GEO: https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE63525

So I downloaded one of the .hic files (Q30) and dumped the matrix as follows:

java -jar ../Software/JUICER/juicer_tools.1.8.9_jcuda.0.8.jar dump -d observed KR GSE63525_IMR90_combined_30.hic 20 20 BP 25000 IMR90_chr20.txt

This generates a symmetric, KR normalized matrix.

Issue: if I provided this matrix as an input to PSYCHIC it won't run and will fail with the error described here:

https://github.com/dhkron/PSYCHIC/issues/5

Now, if I run juicer with

java -jar ../Software/JUICER/juicer_tools.1.8.9_jcuda.0.8.jar dump -d observed KR GSE63525_IMR90_combined_30.hic 20 20 BP 100000 IMR90_chr20.txt

I can execute PSYCHIC successfully on the dataset. What's the difference between the two juicer outputs? Output one has a first column that only consists of 0 entries, because there is no reads within the first 25kb of chr20. That problem is being resolved once I run juicer with a 100kb window.

Is this somehow alarming? I'm trying to understand the difference between the BP parameter I use in the Juicer call and the res parameter in the config file of PSYCHIC.

My second question is: How come there is negative chromosomal coordinates in the .bed files provided on your website? Do these two issues relate somehow and are a result of my lack of understanding the two mentioned parameters?

Thanks,

Soren

[at-ray-o]

Hi, were you able to resolve why there are negative chromosomal coordinates in .bed file? I am having the same issue. Any help would be most welcome.

Thanks, Atreya

[Fuwen-Wang]

My results also have negative chromosome coordinates, can anyone solve this problem？