Reader Comments
Post a new comment on this article
Post Your Discussion Comment
Please follow our guidelines for comments and review our competing interests policy. Comments that do not conform to our guidelines will be promptly removed and the user account disabled. The following must be avoided:
- Remarks that could be interpreted as allegations of misconduct
- Unsupported assertions or statements
- Inflammatory or insulting language
Thank You!
Thank you for taking the time to flag this posting; we review flagged postings on a regular basis.
closeBetween Hive Variance
Posted by physicspolice on 11 Jul 2013 at 22:07 GMT
The list of 300 differentially expressed transcripts was generated by binning larva from 3 exposed vs. 3 control hives.
I wonder how this list compares to a list generated by binning the hives differently?
This would seem a necessary control to determine the statistical significance of the between hive variance.
What if, for example, equally impressive variation in gene expression was found with C3 and IE3 swapped?
Or by swapping any two (or more) hives?
Statistically significant gene expression inside a hive could conceivably be the product of random chance.
It remains to be demonstrated that the exposed hives have expression in excess of this systemic error.
RE: Between Hive Variance
Reinhard replied to physicspolice on 18 Jul 2013 at 20:44 GMT
Uncertainty is inherent to RNA expression data obtained from high-throughput platforms.
Variation between hives and variation within a hive can contribute to altered RNA levels in a given data set. Figure 3 provides a visual indicator that some of this variation indeed exists across treatments.
We agree with the commentator when he/she implies that lists of differentially expressed genes can materialise, simply by comparing any combination of data sets – independent of treatment.
What about our list of 300 differently expressed transcripts that differentiates the C from IE group?
Is this a product of random chance? The result of systematic error? A haphazard assortment of genes with expression levels that happened to pass the statistical filters used to process RNA-Seq data?
Chance occurrence and systematic errors are unlikely explanations for the enrichment of 300 genes operating in lipid-carbohydrate-mitochondrial metabolic networks (Figure 1, (p <1e-8)) and the core glycolytic pathway (Figure 2).