I've aligned several different sets of data for TT2 cells, including ChIP-seq, ATAC-seq and HiC data.
I use bowtie2 as my alignment tool. The data is paired-end reads. The paramters I've used range from default parameters, to "t -q -N 1 -L 25 -X 2000 --no-mixed --no-discordant --no-unal" and more. The reference genomes I used included both mm9 and mm10.
One thing I've noticed is that the alignment rate for TT2 cells is always very low compared to the other cell lines (2-3, J1, DnmtTKO, G7, etc), even when all samples are run on the same lanes on the same flow cell.
Other cell lines typically clock alignment rates of 95%-99% while TT2 alignment rates are usually 50-70%. This holds true across multiple benchwork experiments and dozens of data processing runs.
.
I know this is more of a cell lines question than a bioinformatics question, but has anyone else come across this phenomenon or worked with TT2 cells and noticed some peculiarities?
The bowtie-index is the same for all the different cell lines experiments? Did you manually blast some of the unmapped reads, maybe a contamination in the TT2 stock?
Most samples have been run on both indexes. All samples from the same batch are aligned to the same index(es). I shall BLAST some of the unmapped reads as you suggested. Thank you for the idea. I will update this thread with my results.
Turns out it was a Mycoplasma contamination.