Tweet
Minimum sequencing (and coverage) requirement for human ATAC-seq libraries?
Hi all,
I just got some multiplexed ATAC-seq libraries done, where I got anywhere between 50-90 million reads per sample. Greenleaf's current protocol article (http://onlinelibrary.wiley.com/doi/1...b2129s109/full) suggests that anywhere > 50 mil mapped reads is good to infer open chromatin in human samples, whereas you need > 200 mil reads to footprint transcription factors. We usually do paired end HiSeq 3000 runs.
Can anyone who has done ATAC-seq on human genome provide some insight? What raw read numbers were you shooting for? What kind of coverage is acceptable for publication? How many HiSeq or NextSeq lanes/cells did you end up using?
I'm trying to figure out how much more sequencing we need to do in order to get results that are publishable based on the norms (mostly for coverage) but I'm not too familiar with a lot of ATAC-seq publications yet (besides the primary Greenleaf publications), so if anyone else has an input on this, I would really appreciate it! With the above mentioned metrics, I'm currently at ~3x average coverage for the entire human genome.
Thanks so much!
I just got some multiplexed ATAC-seq libraries done, where I got anywhere between 50-90 million reads per sample. Greenleaf's current protocol article (http://onlinelibrary.wiley.com/doi/1...b2129s109/full) suggests that anywhere > 50 mil mapped reads is good to infer open chromatin in human samples, whereas you need > 200 mil reads to footprint transcription factors. We usually do paired end HiSeq 3000 runs.
Can anyone who has done ATAC-seq on human genome provide some insight? What raw read numbers were you shooting for? What kind of coverage is acceptable for publication? How many HiSeq or NextSeq lanes/cells did you end up using?
I'm trying to figure out how much more sequencing we need to do in order to get results that are publishable based on the norms (mostly for coverage) but I'm not too familiar with a lot of ATAC-seq publications yet (besides the primary Greenleaf publications), so if anyone else has an input on this, I would really appreciate it! With the above mentioned metrics, I'm currently at ~3x average coverage for the entire human genome.
Thanks so much!
Comment