A Neural Network Approach for the Analysis of Reproducible Ribo–Seq Profiles
, Veronica Lachi 2,†, Pietro Bongini 2,3,4,†, Niccolò Pancino 2,4, Monica Bianchini 2,*, Davide Chiarugi 5, Angelo Valleriani 6 and Paolo Andreini 2,†
Round 1
Reviewer 1 Report
The authors employed a statistical method for the identification of highly reproducible Ribo–seq profiles, which was tested on a set of E. coli genes. State–of–the–art artificial neural network models have been used to validate the quality of the produced sequences. New insights into the dynamics of ribosome translation have been provided through a statistical analysis on the obtained sequences. It is an interesting work. But I still have several questions:
1. As a method paper, the authors must provide the data and scripts used in this study. This work must be reproducible.
2. In Figure 1, there are two sub figures. Please add Figure 1A and Figure 1B in the figure legends. And what are the x-axis and y-axis in this figure?
3. In Table 1, the ID means nothing. Please add descriptions of the data.
4. In Table 2, what are the “…”? The rows are not like 1st, 2nd, …. The pattern is not clear. How many rows and columns are included in this table?
5. Figure 4 was just a sequence, not a figure.
6. Figure 6 was poorly made. Please use better template to show the neural network structure.
7. Table 4 was two tables. Please split them into two tables or combine them with revised column names.
8. In Figure 7, the subfigures should add A, B, C, … and explanations in figure legend.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
In this work, machine learning algorithms based on statistical have been implemented for the Analysis of Reproducible Ribo–Seq Profiles. A major revision is required due to several structural issues in manuscript paragraph organization.
· The objective, methodology, and results should be better described, discussed and justified.
· The machine learning methods used in this paper are existing methods, so what are the innovations of this paper?
· It is recommended to add a description of the data part (set of E.coli genes.).
· All diagrams are in poor quality that authors need to redraw/replot and add high quality plots accepted for publishing standards
· Outcome is classification (slow or fast) and you applied SoftMax function. why didn’t you apply Sigmond function on the output layer since this function works in binary?
· I strongly suggest that authors shall carry out more studies to compare the results from this paper to that from other similar studies
· The results should be expanded significantly
· A proof reading by a professional should be conducted to improve both language and organization quality.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors have answered my questions.
Author Response
We thank the reviewer for the comments.
Reviewer 2 Report
Accepts in present form. No comments
Author Response
We thank the reviewer for the comments.
