IGV log scale or auto scale? Which is more precise?

[Huong Nguyen]

Hello,

I am analyzing RNA-Seq profile. I am surprized that the profile is not regular along the gene (up and down, some regions lots of reads, some a few). For each gene, the profile is the same between different replicates. When I compared between 2 conditions, I found interesting differences, which may have biological meanings, in RNA degradation. I was in auto-scale. Then I changed in log-scale (after an advice of bioinformatician), I could not see the differences anymore, so I was a little dissapointed.
So I would like to know log-scale or auto-scale, which is more precise and when we use each one, what are the differences?

Thank you a lot for your help.

Huong nguyen

[James Robinson]

Its impossible to answer this question, both are equally precise.   Log-scale plots your data on a log scale,  so the height of the bar representing the value 10 will be exactly 2x the height of a bar representing the value 1.   In non-log scale it would be 10x as high.    I would refer you back to the bioinformatician that recommended log scale.

--

---
You received this message because you are subscribed to the Google Groups "igv-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/b5a788fd-e872-40d9-add0-143877ae3182%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.