Thanks ATpoint! One more quick Q, do you know how this scaling is performed? Is a random subsample of the larger sample taken? Or are all reads used and then the read counts at each bp are scaled by the 'smaller/larger' ratio?
Confused by MACS2 --scale-to option
Hi all,
Quick Q regarding MACS2. The manual states that the --scale-to option serves to "scale the larger sample up to the smaller sample" or vice-versa. What exactly does it mean by this? Is this scaling only applied when multiple samples/reps are given? Or does it also apply to a sample / control pair?
Context: I currently have an ATAC-seq dataset with a control BAM that has ~ 1/4 as many reads as the sample BAMs - can MACS2 work with this? I've been getting many more peaks without the control than with it.
Thanks in advance for any help! Max
• 1,811 views
•
link
1 answer
It happens when you have -t and -c, then the counts are downscaled to the smaller of the two.
• 0 views
•
link
• 0 views
•
link
Log in to answer this question.