Hello, I have some questions about off-target. If I have understand well off-target can rise from probes that have promiscuous hybridization and recognize more sites in the genome, but also it can rise from libraries that have too large fragments. In this latest case which is the problem if I have probes that are designed adjacent to one another? They capture only the on target and the alignment software should align only using my manifest and not the off target, or there are other problems? In general I know that today the softwares remove the off target, so which is the problem to have it? It decreases the total coverage of the on target? And in this case in which way? Thank you in advance!
Can you be a bit more clear, what exactly is the question? Off-targets will always exist, therefore one aligns the data against the entire genome and then filters for reads spanning the on-target regions.
I would like to understand better which is the problem to have too much off target if than I can remove it; it can decrease the overall coverage? And if there is a way to minimize it? Redisign the probes could solve the problem? And in this case it would be helpful increasing the probe density? Thank you!
I would strictly stick to the established probe design software from the manufacturer as this is probably as optimized as it gets. I do not think it is worth the time and effort to do custom filtering as this would need to be extensively validated and tested to be meaningful, and this will take time and requires experience. The genome unfortunately is redundant or at least similar in many regions so off-targets, depending on the level of sequence similarity between on-and offtarget regions cannot be fully excluded. Stringent alignment parameters will help to avoid false alignments.