Commit ede3913d authored by Carlos GO's avatar Carlos GO
Browse files

carlos edits. maternal edit

parent 54677fe9
......@@ -99,7 +99,7 @@ One of the main conclusions (Discussion, p14) is that intermediate GC content al
\begin{response}{
The effect of GC is seen in Fig 4a, for example, where the yellow curve for GC = 0.5 is the highest. But this could have been shown much more simply by just generating many random sequences with a given GC content and measuring what fraction of these sequences has a multi-branched structure. I am not clear why the number of mutations k is important for this point.}
\resultstag We already performed the experiment suggested by the reviewer in Figure~6 (dotted lines). There, we uniformly sampled sequences and separately plotted the MFE of structures with and without multi-loops (first row), and frequency of multi-branched structures (second row). We note that the average MFEs are comparable and high (i.e. less stable). Moreover, multi-branched structures also appear to not be uncommon in uniform samples.\\
However, it was not clear if the same phenomenon holds true when we focus our analysis to low energy structures (i.e. the ones that are stable enough to carry functions). For this reason, we used \rnamutants to show that sampling shows that the ensemble of low energy structures is enriched with multi-branched structures at GC content bias of 0.3-0.5. Interestingly, it is not the case in the immediate vicinity of random sequences. Instead, mutants have to significantly move away from the initial seed to increase the likelihood to discover stable multi-branched structures.\\
However, it was not clear if the same phenomenon holds true when we focus our analysis to low energy structures (i.e. the ones that are stable enough to carry functions). For this reason, we used \rnamutants to show that sampling shows that the ensemble of low energy structures is enriched with multi-branched structures at GC content bias of 0.3-0.5. Interestingly, it is not the case in the immediate vicinity of random sequences. Instead, mutants have to significantly move away from the initial seed to increase the likelihood of discovering stable multi-branched structures.\\
Variations of the number of mutations k is used to study properties of the local and global neighbourhoods.
\end{response}
......@@ -136,7 +136,8 @@ Section 4.1.2 - R has units of kcal mol-1 K-1 (which are not stated). Then RT =
\begin{response}{
Section 4.1.3 - GC content is maintained by an algorithm associated with mutation. However, this is more similar to selection than mutation. It would be possible to generate new mutant bases with probability controlled by the target GC (theta in the Tamura matrix later in the paper). Then mutation defines the GC content and selection may cause a bias away from theta. The extent to which the observed GC content moves away from theta is a measure of how strong selection is relative to mutation rate. In population genetics it is often interesting to know whether a property arises as the result of mutation or selection. In this model, mutation and selection are not clearly distinguished.
Another funny thing about the algorithm is the while loop. This means that no sequences outside the GC range are ever born. It would be more natural to create sequences by random mutation (for example by the Tamura matrix) and then assign them zero fitness if they are outside the range, so that they cannot be parents for the next generation. The latter seems more reasonable biologically - natural selection cannot tell if something is fit until it exists.}
\claimstag This comment appears to discuss the properties of \maternal, which is not the hypothesis supported in our study. Nonetheless, we remind that we adapted our algorithm from techniques previously used by other groups \cite{Stich2010}.
\claimstag This comment appears to discuss the properties of \maternal, which is not the hypothesis supported in our study. Nonetheless, we remind that we adapted our algorithm from techniques previously used by other groups \cite{Stich2010}. Additionally, while that is a valid suggestion for a GC control algorithm, at high mutational rates and extreme GC contents, there would be a very low probability of obtaining a population with the desired GC content. The reasoning behind our approach would then be to simulate a bias in the nucleotide availability and not a selection bias.
\end{response}
......@@ -190,7 +191,7 @@ Page 2, last paragraph: In which sense are nucleic acids with complex shapes sup
\begin{response}{
Page 3. Comparison with shorter structures has to be done consistently, it is not acceptable to justify the interest of looking for multi-branched structures referring to works that analyzed shorter sequences and in consequence did not find them.}
\hypothesistag We are puzzled by this statement. We have no claim saying that we identified multi-branched structures that were not found by previous studies on shorted sequences. As stated in our manuscript \emph{``This approach considerably expands the scope and significance of comprehensive RNA evolutionary studies that were previously limited to sequences with less than 20 nucleotides''} (page~4) or later \emph{``Our analysis completes recent studies that aimed to characterize fundamental properties of genotype-phenotype maps''} (page~22). Knowing that complex structures occur on larger sequences, it is logical to investigate properties of bigger sequence-structure maps.
\hypothesistag We are puzzled by this statement. We have no claim saying that we identified multi-branched structures that were not found by previous studies on shorter sequences. As stated in our manuscript \emph{``This approach considerably expands the scope and significance of comprehensive RNA evolutionary studies that were previously limited to sequences with less than 20 nucleotides''} (page~4) or later \emph{``Our analysis completes recent studies that aimed to characterize fundamental properties of genotype-phenotype maps''} (page~22). Knowing that complex structures occur on larger sequences, it is logical to investigate properties of bigger sequence-structure maps.
\end{response}
\begin{response}{
......@@ -236,7 +237,7 @@ As explained earlier, we verified that our results are fully reproducible w.r.t.
\begin{response}{
Figure 4 is also confusing. Are panels b), c) and d) including all sequences probed, at any mutational distance? How can the complete set of multi-branched structures be compared? Again, it is essential to demonstrate that the quantitative results in Figure 4 are independent of the number of sequence-structure pairs sampled.}
\resultstag All figures include all sample data. As it is often the case when analyzing multi-dimensional data, we projected the data on different subspaces to offer a complete picture of their distribution. Although, We are puzzled by this comment since the panel b) already include the mutation dimension.\\
\resultstag All figures include all sample data. As it is often the case when analyzing multi-dimensional data, we projected the data on different subspaces to offer a complete picture of their distribution. Although, we are puzzled by this comment since the panel b) already include the mutation dimension.\\
Again, as explained earlier, we verified that our results are fully reproducible w.r.t. the number of samples.
\end{response}
......@@ -253,7 +254,7 @@ However, we are puzzled by this comment which seems to mix remarks on \rnamutant
\begin{response}{
Last paragraph. It is stated that their algorithm fails to generate the structural complexity found in real populations?. Which real populations are the authors comparing with? A single one? The complete database? Is this comparison sensible at sufficiently long evolutionary times (natural versus synthetic populations)?}
\resultstag We assume that the reviewer is discussing the results of the \maternal algorithms, which is not the hypothesis we are supporting. As described in the introduction, we compare our results RNA families from the Rfam database \cite{Nawrocki:2015aa}, arguably one of the most popular database in RNA bioinformatics. We also ran our simulations on long evolutionary time to guarantee the robustness of our results (E.g. Figure~S2).
\resultstag We assume that the reviewer is discussing the results of the \maternal algorithms, which is not the hypothesis we are supporting. As described in the introduction, we compare our results to RNA families from the Rfam database \cite{Nawrocki:2015aa}, arguably one of the most popular databases in RNA bioinformatics. We also ran our simulations on long evolutionary time to guarantee the robustness of our results (E.g. Figure~S2).
\end{response}
\begin{response}{
......@@ -273,7 +274,7 @@ Page 14. Two interpretations are given to the presence of these structures, and
\begin{response}{
Page 15. I definitely disagree with the conclusion that "Variations of the sizes of populations or lengths of RNA sequences (...) we do not expect any major impact on our conclusions" for all the reasons exposed.}
\claimstag We believe that there is a profound misunderstanding here, for which we take full responsibility. First, as mentioned above, we consider various population sizes to guarantee robustness of our results. Next, we were only speaking of minor variation of the sequence length corresponding to a couple of insertion or deletion. As the reviewer can note by looking at the citation at the end of the sentence, this remark is mainly to introduce the availability of more sophisticated algorithm modelling indels \cite{Waldispuhl:2002aa}.
\claimstag We believe that there is a profound misunderstanding here, for which we take full responsibility. First, as mentioned above, we consider various population sizes to guarantee robustness of our results. Next, we were only speaking of minor variation of the sequence length corresponding to a couple of insertions or deletions. As the reviewer can note by looking at the citation at the end of the sentence, this remark is mainly to introduce the availability of more sophisticated algorithms modelling indels \cite{Waldispuhl:2002aa}.
\end{response}
\begin{response}{
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment