Profound Non-Randomness in Dinucleotide Arrangements within Ultra-Conserved Non-Coding Elements and the Human Genome
Round 1
Reviewer 1 Report
Comments for author File: 
Comments.pdf
Author Response
Our reply is attached
Author Response File: Author Response.pdf
Reviewer 2 Report
Considering next neighbour distances of DN as field of investigation is an interesting and good approach.
line 32
"significant (>10%) non-random genomic arrangements"
you specify in line 160 how you get 10% and also refer later in Table 2 to abs(RPD) threshold
in the abstract it is not clear what you mean by this ">10%" and how you obtain it
Line 50
"and the rest of the genome"
according to reference 4 this has been based on a comparison of UCNEs to whole rest of the genome but not on individual subsets as CpG islands
is there any information available on such tiny subsets of the genome in which these features might also be present but would not become visible if analysed with all "the rest of the genome"?
line 144
you reduce the standard deviation (sigma) by sqrt(1000) and you get SE
table 1
what is "av" in 3-5?
would a switch in 6-8 of first and second line be better
(WG in first line and SD in second?)
line 175
WH?
line 181
"only in the forth"
ambigous, as it is the forth letter of DN GpC plus DN TpA, but it is L=4 and so it would be GCNNTA vs. GCNNTT and so would be sixth
line 192
you describe
relative percentage difference (RPD)
but you already use it in line 156
line 218 and 219
pick?
Table 3 and Figure 3
RAND is always respective RAND of set given?
so UCNE vs. RAND is UCNE vs. RANDUCNE?
Figure 4
order in legend may be changed based on taxonomy or other order
line 270
reference for the B-DNA?
line 292
often exceeds 10% and even reaches 40% in ex-treme cases
may be you add some info based on RPD
line 298
distance L=4 corresponds to a half turn
it puts your DN in the center of a half turn after completing 5bp (one nt before and one after the half)
and so you come up to L=3 to L=5 relating your DN to a half turn
(maybe just a few sentences for explanation)
line 302 pick
line 315
you may use AI if you want, it is not necessary
on what will you train?
line 317
why now just mammals?
you now try to focus on the relevance of DNs...you may add just a sentence before this paragraph to smoothen it for the reader and explain why you do focus
line 388
is less than 20% bigger of
than?
comes not
does not...
line 392
may you explain with a few sentences why this is sufficient for yor testing?
and why now 100?
very few errors
spelling
article
Author Response
Our reply is attached
Author Response File: Author Response.pdf
Reviewer 3 Report
The paper titled 'Profound Non-Randomness in Dinucleotide Arrangements within Ultra-conserved Non-Coding Elements and the Human Genome' by Fedorova and co-authors presents an intriguing exploration of a complex genomics issue. The authors discovered a unique pattern in non-coding sections of Ultra-conserved Non-Coding Elements (UCNE). Their approach is well-executed, but one limitation is the lack of direct connection with experimental data.
In the discussion, the authors suggest that : ".DN local non-randomness and specific distributions likely play important oles in chromosomal assembly and functioning.". To further validate their findings, it might be worthwhile for the authors to explore publicly available data, such as ChipSeq or ATAC-seq, which could complement their analysis.
While UCNEs are primarily discussed in vertebrates, the inclusion of arthropods in the study introduces an interesting perspective. The slight variation observed in Drosophila could be attributed to the differences in GC content, which is known to vary among arthropods. To bolster this finding, it could be beneficial to compare the results with other well-known arthropod model organisms, such as Tribolium castaneum and Apis mellifera. Otherwise the necessity of including Drosophila in the paper is not quite clear.
Author Response
Our reply is attached
Author Response File: Author Response.pdf
