search

jianhong / ATACseqQC

ATAC-seq Quality Control
https://jianhong.github.io/ATACseqQC/articles/ATACseqQC.html
23 stars 12 forks source link

splitBam fragment sizes for nucleosome-free DNA #42

Open rauldiul opened 3 years ago

rauldiul commented 3 years ago

Hi,

I'm using your package and have a general question regarding the sizes in which BAM file is split with splitBam to generate nucleosome-free, mononucleosome... fragments. I see that the size limits for splitting the BAM are similar to Buenrostro et al. 2013. However, why is there a 38 bp lower limit for nucleosome-free fragments? Why are not nucleosome-free fragments simply specified as < 100 bp? I went through the ATACseqQC paper but could not find a justification for this.

thanks for your help!

etiennedanis commented 3 years ago

From what I read (Adey et al. Genome Biol. 2010), it looks like the Tn5 transposase cannot generate anything less than 38bp. So any fragment shorter than 38 bp should just be artifacts.

rauldiul commented 3 years ago

Great, thanks for the feedback and the reference!