General comments:
The revised manuscript entitled “Multi-platform microRNA profiling of hepatoblastoma patients using formalin fixed paraffin embedded archival samples” has addressed several concerns from the initial submission, especially with regard to adding value over the previously published Scientific Reports paper. However, a few items should still be addressed before the article is appropriate for publication.

Major Compulsory Revisions:
1) The authors added several new tables and analyses to the manuscript, including additional clinical data, qRT-PCR data, quality metrics, and commentary about starting material needed for each assay. The comparison with previous studies, however, could be clarified and improved upon. Overall the text in this new section does not adequately describe the analysis conducted (for example, the P-value associated with the hypergeometric test will depend strongly on what is used as a reference set of miRNAs – is it the set interrogated on the array, nanostring, or all of miRbase?). It also isn’t clear why the authors compared their detected miRNAs to those differentially expressed in a different cancer type; it has been shown that miRNA expression patterns are largely tissue-specific, so this overlap may not be meaningful. Finally, the comparison between up-regulated and down-regulated miRNAs from the Magrelli dataset is difficult to interpret given the lack of normal controls and possible within-cancer heterogeneity. Given the inherent challenges of conducting a statistically sound integration of these datasets, I would suggest simply reporting the ability to detect previously identified differentially expressed miRNAs. This would highlight the ability to measure potential miRNA drivers of tumor phenotypes within hepatoblastoma using the new dataset and suggest another way for the data to be used.

Minor Essential Revisions:
1) The image resolution on Figure 2 is too low (lines and text are not sharp at 100% magnification).
2) The clinical data are not entirely clear. Relapse-free-survival and overall survival are both appropriate data to include, but there should also be a vital status indicator (i.e. dead or alive). In its current form one cannot tell whether patients died from a recurrence diagnosed post-mortem, versus being alive at last follow-up and censored for survival analysis.
Level of interest An article of importance in its field
Quality of written English Acceptable
Statistical review No, the manuscript does not need to be seen by a statistician.
Declaration of competing interests I declare that I have no competing interests.

Authors' response to reviews: (http://www.gigasciencejournal.com/imedia/1308694195192957_comment.pdf)

The reviewed version of the manuscript can be seen here:

All revised versions are also available:
First revision -