This is a valuable contribution to the technical development in the mutagenesis to aid genetic analysis of non-model organisms and especially evolutionary study of genetics and phenotypes. Molecular biology can be benefitted from this work too. These authors have computationally predicted the genomic distribution of the target sites for TALEN and CRISPR in a number of organisms whose genome sequences are available. So the database for these target sites will be consulted for any researchers who want to pursue the technical analysis in these organisms. The discussion about the two techniques is informative, not redundant with previous technical description. The ms, after revision for a number of compositional problems, should be published. Most of these problems are minor, with a few places that include erroneous understanding or unclear thinking, can be easily fixed. I gave some possible ways for revision and they are welcome to incorporate all the suggested changes.

There is one thing unclear after reading the ms: are the actual databases for the predicted target sites going to be published with this paper? I would strongly encourage them to do so, with it publically available, the work would be of potentially big impact.

• Page 3: Paragraph 1 – when citing for knockdown, other non-mammals, e.g. fruitflies and nematodes, should be included.

• Page 3: Paragraph 1 – “labor-insensitive,”: what does it mean?

• Page 3: Paragraph 2 – “Last year” ⇒ “Recent years”, because many cited appeared before last year.

• Page 4: Paragraph 1 (and abstract) – “…exciting prospect for disclosing “black boxes” in the near future.” The term “black boxes” should be explained and here is unclear what it is meant.

• Page 4: Paragraph 2 “we will not discuss them in detail here, but instead, summarize the frame of their mechanisms as shown in Figure 1.” A more reasonable statement would be something like: “We will not prepare a redundant review but first provide a summary of the actual mechanisms for further discussion….”

• Page 4: Paragraph 2 “the two key methods of DNA repair mechanisms in eukaryotic cells”: delete “methods of” to be more concise and accurate.

• Page 4: Paragraph 2 “the 3-nucleotides recognizing model, the system is also very expensive and difficult to assemble.” Delete “very”.

• Page 5: Paragraph 2: “in/dels” ⇒ “indels”, which is more conventionally types. “drosophila” ⇒ “Drosophila”

• Page 6: Paragraph 1: “wheat”: is or isn’t it a model organism, given so many people worked on it in so many decades? The polyploidy crop plant has been taken as model to understand evolution of chromosomes, although molecular biologists are not caring it but its genetic analysis has been attractive to scientists of generations.

• Page 7: Paragraph 1: “Given those functional experiments, it indicates that new genes may play critical roles in essential biological processes”: there are several rigorous experiments in fruit flies and nematodes should be cited: for example (not completely listed here), Ross et al (2013, Science. 340, 1211); Ragsdale et al (2013, Cell. 155, 922); Chen et al (2012, EMBO J, 31, 2798). In addition, to stick to “essential” function is not the point the area should be focused only and the real issue is “do new genes play important roles in function and phenotypic evolution” so any new genes that have detectable functional or phenotypic effects are important for evolution and biology of genes. So I would suggest first cite the relevant literature above, and then change this statement, somewhat misleading in the context, into something like: “Given those functional experiments, it indicates that new genes may play significant roles in important biological processes or phenotypes”. This would be leading audiences to correctly understand the issues of new gene studies.

• Page 7, Paragraph 1: “Therefore, the exact functions of a large amount of new genes are still unknown. Large-scale functional studies on new genes by RNAi are reported, but turned out to be unstable due to its incompleteness and unpredictable off-target effects [60].”

These two statements do not make the points the discussion here wants to make, although I understand their enthusiasm for what they are discussing. For the first: There is nothing we can call “exact function(s)” of genes in nature, as this review is caring, for a very general reason that has been realized long time ago by many biologists: no artificially control “experiment” can reveal “exact” functions because controls are what we have to set to make inferences and which are artificial; secondly, almost every gene has multiple functions, called pleotropic depending upon how we define.

The second statement is also not addressing the real point and neglecting productive practice in Drosophila genetics in recent decade since the technique is used widely in the field, as is doing today. The consensus in the community is that RNAi and mutagenesis are two complementary techniques and each has both its disadvantage and advantage over the other. For example, the mutagenesis would have greater challenge than RNAi to study lethal phenotype because the later can be more easily to assemble conditioning operation. Further, the “unstable due to its incompleteness and unpredictable off-target effects” is not worried by most in the field because the technique can get improved, including the control of “off-target effects”. So, the statements can be revised in more reasonable and objective ways like:

“Therefore, the relevant functional information of large amount of genes, including new genes, is to be explored. Large-scale functional studies on new genes by RNAi are reported, providing candidates with important functions for further study using complementary techniques such as the mutagenesis approaches discussed in this review [60].

• Page 8: “Humans have domesticated hundreds of plants and animals species as sources of food and materials over the past 12,000 years [71].” It is likely earlier than 12,000 years, given the old rock painting that recorded some animals that has been domesticated. Check out the literature, which focused on crop plants.

• Page 9: the evidence from GWAS is not much convincing in general, although it is easier to do in a programming fashion. I would suggest to give some “buts” to this line of evidence.

• Page 10: “revolutionize”: many practitioners of CRISPR complained it was not so easy to use as it is advertised. So to be careful not to use the word of “revolutionize” easily. In this statement, “promising” is a good word to use.

• Page 11 and 6: “PAM” should be defined immediately (PAM, protospacer adjacent motif). I found this from their table 2 later

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.