Comments on abstract, title, references 1) In the abstract, the authors described the background, aim, and all the major findings of the work. However, they do not mention the approach they took to investigate the research aim. 2) The authors stated the take-home message of the work aptly in the title. 3) The authors profusely studded the article with all the relevant and key references. However, key articles can be cited in a few places. For example, when describing the third result section (related to Fig. 5) the authors may reuse some important references like Gómez-González et al 2019, and Hill et al 2016. In fact, data from Fig. 5 reinforces the earlier findings from Hill et al 2016. The authors could also consider mentioning the work by Gianni M et al 2020 (previously as a preprint and now published) that describes similar findings as this work.

Comments on introduction 1) The authors mentioned both the general and specific backgrounds relevant to the current study. 2) They clearly based their research question about the role of TDP-43 in regulating R-loops on previously published articles. 3) They wrote a cohesive introduction introducing the broader topic of R-loops and their role in genome stability and how RNA binding proteins (here TDP-43) orchestrates these processes. 4) However, they could also mention very briefly how RNA binding proteins regulate R-loop levels and how this is relevant for the current study. This could be especially important for readers who are not familiar with the context.

Comments on methodology 1) The authors reported all the materials and methods in great detail that can be used by other researchers. 2) They also provided details of the number of individual experiments, replicates, and statistics used in the current study. 3) However, in my understanding, the authors used three different kinds of siRNAs to deplete TDP-43 in three cell types: HeLa, SH-SY-5Y, and mouse primary neurons. As the first two cell types are of human origin, they could have picked a different TDP-43 siRNA for the mouse primary neurons. But the TDP-43 siRNA was only able to partially reduce TDP-43 levels in mouse primary neurons (Fig S7). In any way, it is advisable to perform the key experiments using at least three different pools of siRNAs in all the cell lines used in the study. 4) To assess the chromosome aberrations (related to Fig 2D), the authors wrote in the methods section as follows: “Chromosomes were assessed for both chromatid breakage and end-to-end fusions. For each condition, the total number of aberrations counted was divided by the total number of metaphase spreads assessed to determine the average number of aberrations per cell.” This might not be an apt way to evaluate chromosome aberrations because making a gross percentage of chromosome aberrations in all the metaphases could give insufficient data. Instead, the authors could consider representing the data by evaluating the number of chromosomes present in one metaphase and how many such metaphases exist in a given sample. 5) Lastly, the authors should also mention if they blinded the image analysis related to this study.

Comments on data and results 1) The data is presented appropriately and directs to the relevant figures in the text. 2) The authors clearly stated wherever the data is statistically significant. 3) The authors mostly used the regular conventions in the field to measure different parameters. However, in two cases the following changes can be made to represent the data in an efficient way. 3.a) Fig 2D: The authors could score the chromosomal aberrations per metaphase with a cut-off. For example, metaphases with ≥ 5 aberrations can be scored per condition. This would provide better information for the readers than the percentage of cells with chromosomal aberrations. 3.b) Fig 3A, 5B, 6D: The authors scored the S9.6 intensities to evaluate the nuclear R-loops levels using IF. It is good that they used a non-nucleolar signal for the study. However, they normalized the signal to the siRNA control cells to report the “standardized S9.6 mean integrated density”. Although this is informative to some extent, an appropriate way to do this is to simply plot the raw intensities without any normalization as a box plot.

Comments on discussion The authors summarized the key findings of the study and discussed them cohesively by drawing conclusions and models from varied sources. They have stated the limitations of the study without going overboard to make claims beyond the reported data. They proposed a molecular model with which TDP-43 could induce pathological R-loops and how this is relevant to disease-causing TDP-43 mutations in neurodegenerative diseases. They compared their data with other similar studies and provided alternative explanations wherever necessary. The authors could extend their work and propose future directions. For example, they noted how anomalies in R-loop and TDP-43 levels are associated with neurodegenerative diseases. They can extend this briefly and propose how their work could potentially bring about therapeutic strategies to alleviate complex neurodegenerative diseases.