blacklist annotation bin incorrect?

[ypradat]

Dear Aziz,

I am interested in adding another track for my own normalization purposes and first I am trying to understand how the current annotation bins were generated.
I looked at the code in the README and ran it for the 50 kbp window.

I tried recreating the blacklist annotation as I did not understand the values provided in the package (for example binSize=50, genome=hg38).

I ran

> wget https://github.com/Boyle-Lab/Blacklist/blob/master/lists/hg38-blacklist.v2.bed.gz?raw=true -o hg38-blacklist.v2.bed.gz
> gzip -d hg38-blacklist.v2.bed.gz

> R
> bins <- getBinAnnotations(binSize=binsize, genome="hg38", type="SR50")                    
> bins <- bins@data                                                                         
> bins$blacklist_check <- calculateBlacklist(bins, bedFiles=c("./hg38-blacklist.v2.bed")) 

and got

>   head(bins %>% select(chromosome, start, end, gc, mappability, blacklist, blacklist_check))
                chromosome  start    end       gc mappability blacklist blacklist_check
1:1-50000                1      1  50000 48.42500    19.11770         0             100
1:50001-100000           1  50001 100000 37.67400    23.19640         0             100
1:100001-150000          1 100001 150000 42.99600     6.34175         0             100
1:150001-200000          1 150001 200000 48.63400    10.00230         0             100
1:200001-250000          1 200001 250000 48.05635    15.25810         0             100
1:250001-300000          1 250001 300000 39.14446     8.19869         0             100

As you can see, black_list_check and blacklist are different. Looking at the blacklisted regions of chr1,

> cat hg38-blacklist.v2.bed| awk '$1=="chr1"'
chr1    0       792500  High Signal Region
chr1    91386300        91388400        Low Mappability
chr1    103594400       103760600       High Signal Region
chr1    121605200       124938900       High Signal Region
chr1    125067600       125086000       High Signal Region
chr1    125130200       143562200       High Signal Region
chr1    161423100       161472400       High Signal Region
chr1    168348600       168349900       High Signal Region
chr1    224010800       224017000       High Signal Region
chr1    236713000       236715600       Low Mappability
chr1    248932700       248956400       High Signal Region

It makes sense that all first bin regions are annotated blacklist=100% as they completely overlap the blacklisted region chr1:0-792,500. Could you please clarify the reason for the discrepancy ?

Best,
Yoann

[asntech]

@ypradat Thank you for reporting this, and sorry, I saw this late. It appears we've used an older version of the blacklist bed file in the current version.

Can you please give a try to the updated version in the dev branch?

remotes::install_github("asntech/QDNAseq.hg38@dev")