Reader Comments

Post a new comment on this article

too short?

Posted by emptyhb on 21 Dec 2011 at 16:34 GMT

averaged ∼400 bp
http://ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1002256#article1.body1.sec3.sec3.sec4.p1

I downloaded dataset1 (sup material) and did a histogram of the length of predicted regulatory elements. To my surprise, the length is at most 400 bp, where the majority fall between 250-350. Is this a result of the prediction algorithm or is it limited by the pcr cloning step? Critically, we know that a minimal enhancer in Drosophila may be well beyond 400 bp (like eve stripe 2), therefore I'm worried that this short length might contribute to an overall low validation rate in-vivo, as you are constrained to find those extremely short enhancers, which are very likely uncommon.

No competing interests declared.

RE: too short?

Wyeth_Wasserman replied to emptyhb on 13 Feb 2012 at 04:55 GMT

Due to the nature of the assay used, we needed to keep the insert sizes within a narrow range. (Otherwise the selection of clones from the library would bias to shorter inserts.) For vertebrate enhancers, there is sufficient evidence that the size included should mediate tissue-specific enhancer activity in a cell culture assay. While longer sequences often can produce quantitatively higher expression, the core sequences can be expected to mediate a detectable tissue-specific expression. It is a fair point for discussion, and it would be interesting to see if longer sequences perform differently in a similar assay.

No competing interests declared.