The manuscript by Jin and colleagues reports the chromosome-scale assembly of the genome of the river prawn Macrobrachium nipponense, an evonomically important crustacean species, and investigates potential sex-related candidate genes which might serve as potential molecular markers for early sex determination. In summery, the quality of the reported genome assembly was good, thanks to the combination of PacBio, Illumina PE and Omni-C libraries. However, aside from the availability of this important resource for the aquaculture sector, I found several other aspects of this MS rather weak. First, from a methodological point of view, it was quite difficult to ascertain whether all gDNA libraries were obtained from a single individual (as it should) or not. Second, the authors built a phylogenetic tree with a phylogenemic approach, which however lacks several methodological details and a few key crustacean species that should have been included. In particular, one of the key findings of this study (a WGD event that occurred in M. nipponense) has not been discussed at all, and its timing has not been investigated in sufficient detail, leaving the reader with more doubts than before. The authors did not report whether this was an expected finding and did not discuss this important data in relation with previously published literature, including karyotype studies carried out in multiple Macrobrachium species and cytogenetic estimates of c-value. I'm also not fully convinced by the DEG analysis, as the authors should have tried to put more emphasis on the reliability of the candidate genes identified, which would have been expected to be characterized by very high fold-change values. This type of information are unfortunately not available in the present version of the manuscript, which merely reports the enrichment of KEGG terms, with little attention to the fact that such terms are strongly biased towards model species. To discuss such data in a more reliable way, the authors should have considered the enrichment of GO terms and conserved domains as well. Overall, I do not feel this work provides at the present time data of sufficient novelty or depth of investigation to be published on this journal. I would strongly suggest the authors to consider with much more attention the WGD event, as this could bring this work additional value. Alternatively, considering the good quality of the genome assembly itself, this could easily be published as a short genome assembly note on another journal.

L48: approximate - approximately

L63: some reads are here referred to the company name (e.g. PacBio), others to the platforms name (Hiseq), others to the library type (Hi-C). Please be more consistent: PacBio is fine (you may use Pacbio long reads), but replace the other two terms with "paired-end and Hi-C libraries processed on an Illumina platform".

L69: why did the authors use multiple individuals to obtain a reference genome? In principle, it would be always preferable to extract gDNA from a single individual, in order to minimize the impact of polymorphisms in the assembly, especially when expected heterozygosity is high.

L76: Some information is missing here, with the regards to the multiple individuals sampled. In particular, were the libraries generated through pulled samples from the five individuals, or where different individuals used to generate different libraries?

L108: this requires an important clarification. Sequencing data from two different libraries have been used here, but as I mentioned before it is unclear whether these two libraries have been obtained from pooled gDNA from multiple individuals or from single individuals. In the former case, a correct estimate of genome size (and heterozygosity) is not possible.

L122: polish -> polishing

L123: I would have expected to see some information concerning Hi-C library preparation and sequencing, which are not present here.

L152: "Ostreae Concha, Fucata martensii". There is an error here, as these are not correct scientific names. The authors probably refer to Pinctada fucata martnesii, and I don't know what Ostreae Concha is. If this is not a species, then the species count would be eight, not nine. Also penaeus should be Litopenaeus.

L164: "by aligning against". Please be more specific, as this is also a BLAST-based analysis.

L166: again, this is not an alignment, but rather a detection of HMM models

L185: several details are missing here, including the molecular model of evolution set 8and a description of how it has been selected).

L209: DEGs: which comparisons were investigated here? I guess between the two seasons, but please be more specific here.

L210: folds should be "fold change"

L221: missed -> missing

L248: "four iag genes" -"four paralogous iag genes"?

L249: rather than the number of genes, it would be much more interesting here to get to know the distance in MB between the 3 genes.

L291: "lower organism"???

Discussion: I found it very disappoint to find no mention at all about the predicted WGD event. This interesting finding is only mentioned in the results section, and not discussed at all. Was this expected? Are other WGD events known in Palaemonidae or Pleocyemata? How do the number of assembled choromosomes and genome size compare with cytogenetic estimates? I see, from the animal genome size database, various c-value estimates from different Macrobrachium species, ranging from 6.48 to 22.16 (but data from M. nipponense is not available).

A number of studies exist concerning Macrobrachium karyotype. According to Damrongphol and colleagues (1991), the aploid number of chromosomes in M. rosenbergii is 59, and this value is confirmed in M. malcompsonii (Rebecca et al. 2020). M. lachesteri has n = 58 (Phimphan et al. 2018). Others, like M. acanthurus and M. amazonicum have n=49 (Molina et al. 2020), as reported here, but this should be mentioned in the manuscript. More importantly, is there any clear evidence that Macrobrachium spp. has undergone a WGD event? I think this should be supported by much more substantiated evidence, which would allow a more precise timing for this event. Honestly, the closest genomes included in this analysis were those of L. vannamei and P. virginalis, but these species are Pleocyemata and quite distantly related with M. nipponense. Why were not the genomes of Pandalus platyceros, Caridina multidentata and Palaemon carinicauda used, since these are evolutionarily closer and available?

Table 2 is incomplete. What do these numbers represent? Fold change values?

Declaration of competing interests Please complete a declaration of competing interests, considering the following questions: Have you in the past five years received reimbursements, fees, funding, or salary from an organisation that may in any way gain or lose financially from the publication of this manuscript, either now or in the future? Do you hold any stocks or shares in an organisation that may in any way gain or lose financially from the publication of this manuscript, either now or in the future? Do you hold or are you currently applying for any patents relating to the content of the manuscript? Have you received reimbursements, fees, funding, or salary from an organization that holds or has applied for patents relating to the content of the manuscript? Do you have any other financial competing interests? Do you have any non-financial competing interests in relation to this paper? If you can answer no to all of the above, write 'I declare that I have no competing interests' below. If your reply is yes to any, please give details below.

I declare that I have no competing interests.

I agree to the open peer review policy of the journal. I understand that my name will be included on my report to the authors and, if the manuscript is accepted for publication, my named report including any attachments I upload will be posted on the website along with the authors' responses. I agree for my report to be made available under an Open Access Creative Commons CC-BY license (http://creativecommons.org/licenses/by/4.0/). I understand that any comments which I do not wish to be included in my named report can be included as confidential comments to the editors, which will not be published. I agree to the open peer review policy of the journal.

Authors' response to reviews:

Reviewer #1: The manuscript by Jin and colleagues reports the chromosome-scale assembly of the genome of the river prawn Macrobrachium nipponense, an economically important crustacean species, and investigates potential sex-related candidate genes which might serve as potential molecular markers for early sex determination. In summary, the quality of the reported genome assembly was good, thanks to the combination of PacBio, Illumina PE and Omni-C libraries, and this will undoubtedly represent an important resource for the aquaculture sector. However, as a data note, I believe this work tried to include way more biological data than it should have, considering that such data (in particular the data concerning WGD) has not been appropriately analyzed and discussed. While my detailed comments are appended below, I anticipate that I would strongly suggest the authors to reshape this MS as a data note, i.e. by purging the text from most of the parts linked with sex determination candidates (which remains, in my opinion, rather weak) and WGD (whose possible existence may be briefly mentioned). Reply: Thanks for your instructive comment. Yes, we reshaped our manuscript in accordance with your advice. That is to say, we limited the descriptions of DEGs to a bare minimum (Line 301-311), and expanded the discussions of WGD and karyotype comparisons of Macrobrachium species in the Discussion section （lines 287-300）.

From a methodological point of view, it was quite difficult to ascertain whether all gDNA libraries were obtained from a single individual (as it should) or not. This is a key technical issue that needs to be solved first and foremost, as it may possibly affect the quality of the genome assembly itself. Reply: You are right, gDNAs are usually extracted from a single individual for whole genome sequencing so as to minimize the adverse effects of polymorphisms. However, the pooled muscles from one specimen of our river prawn weighted up to 4 g, although the total body weight individually was at a range of 13.02 to 15.56 g. It is not sufficient to extract enough gDNAs from a single individual for the whole genome sequencing project. Thus, we had to pool the muscle tissues from 5 individuals for the practical works. These five prawns were born by the same parent pair and then cultivated by us in the same pond of our local aquaculture base. This is a popularly compromise way for small animals. Your reconsideration is appreciated.

The authors built a phylogenetic tree with a phylogenemic approach, which however lacks several methodological details and a few key crustacean species that should have been included. This type of analysis would most certainly not be sufficient to be included in a full paper, and it is not necessary in a data note article. Reply: We quite agree with you to use more crustacean species for a better phylogenetic tree. We considered that a chromosome-level genome assembly with continuous scaffolds would be more appropriate for a comparative phylogenomic analysis. However, those key species you recommended are reported with low-quality assemblies. For example, Pandalus phatyceros (scaffold N50: 1,512 bp; NCBI accession number: GCA_005815305.1), Caridina multidentata (scaffold N50: 819 bp; GCA_002091895.1), and Palaemon carinicauda (scaffold N50: 962 bp; GCA_004011675.1/) are too fragmented to be used for this phylogenetic analysis. By the way, more methodological details were provided in lines 195-197 for your reconsideration.

One of the key findings of this study (a WGD event that occurred in M. nipponense) has not been discussed at all, and its timing has not been investigated in sufficient detail, leaving the reader with more doubts than before. The authors did not report whether this was an expected finding and did not discuss this important data in relation with previously published literature, including karyotype studies carried out in multiple Macrobrachium species and cytogenetic estimates of c-value. Reply: Thank you for your nice advice. Yes, it is done. Related discussions of WGD and karyotype studies were provided in the revised manuscript. Please see more details in lines 287-300 under the Discussion section.

I'm also not fully convinced by the DEG analysis, as the authors should have tried to put more emphasis on the reliability of the candidate genes identified, which would have been expected to be characterized by very high fold-change values. This type of information are unfortunately not available in the present version of the manuscript, which merely reports the enrichment of KEGG terms, with little attention to the fact that such terms are strongly biased towards model species. To discuss such data in a more reliable way, the authors should have considered the enrichment of GO terms and conserved domains as well. Again, discussing such aspects in a full paper would require much more attention, and the current presentation of these data is excessive for a data note article. I would suggest the authors to limit the reporting of DEGs to the bare minimum, stating that a few plausible sex-related gene candidates have been identified, but that these will require further independent validation. Reply: Thanks for your instructive comments. According to your advice, we tried to limit the descriptions of DEGs to a bare minimum (Line 301-311), and stated more about a few sex-related candidate genes although these primary conclusions require in-depth validations. See more details in lines 272-284 and 307-311.

L48: approximate – approximately Reply: Sorry for the mistake. Yes, it is done in line 47.

L63: some reads are here referred to the company name (e.g. PacBio), others to the platforms name (Hiseq), others to the library type (Hi-C). Please be more consistent: PacBio is fine (you may use Pacbio long reads), but replace the other two terms with "paired-end and Hi-C libraries processed on an Illumina platform". Reply: Thanks for your nice advice. Yes, it is done. We revised this sentence as follows in lines 61-63. In our present study, a chromosome-level genome assembly of the Oriental river prawn was constructed by integration of Pacbio long reads, Illumina short reads, and Hi-C sequencing data.

L69: why did the authors use multiple individuals to obtain a reference genome? In principle, it would be always preferable to extract gDNA from a single individual, in order to minimize the impact of polymorphisms in the assembly, especially when expected heterozygosity is high. Reply: You are right, gDNAs are usually extracted from a single individual for whole genome sequencing so as to minimize the adverse effects of polymorphisms. However, the pooled muscles from one specimen of our river prawn weighted up to 4 g, although the total body weight individually was at a range of 13.02 to 15.56 g. It is not sufficient to extract enough gDNAs from a single individual for the whole genome sequencing project. Thus, we had to pool the muscle tissues from 5 individuals for the practical works. These five prawns were born by the same parent pair and then cultivated by us in the same pond of our local aquaculture base. This is a popularly compromise way for small animals. Your reconsideration is appreciated. See more details in lines 77-79.

L76: Some information is missing here, with the regards to the multiple individuals sampled. In particular, were the libraries generated through pulled samples from the five individuals, or where different individuals used to generate different libraries? Reply: We pooled multiple individuals to generate different libraries. Thus, we revised this sentence as follows in lines 77-83. Five individuals from each group were pooled, and muscle DNAs from pooled samples were extracted using a Nucleic Acid Kit (Qiagen, Germantown, MD, USA) in accordance with the manufacturer’s instructions. The extracted gDNAs was then used for constructing libraries for Illumina (Illumina Inc., San Diego, CA, USA) and PacBio (Menlo Park, CA, USA) sequencing. According to the Illumina’s instructions, seven paired-end libraries were constructed with the following insert sizes: 270 bp, 500 bp, 800 bp, 2 kb, 5 kb, 10 kb and 20 kb.

L108: this requires an important clarification. Sequencing data from two different libraries have been used here, but as I mentioned before it is unclear whether these two libraries have been obtained from pooled gDNA from multiple individuals or from single individuals. In the former case, a correct estimate of genome size (and heterozygosity) is not possible. Reply: The pooled muscle tissues from individual prawn were up to 4 g, which is insufficient to extract enough gDNAs for the whole genome sequencing project. Thus, we pooled multiple individuals to generate different libraries. In fact, this compromise way has been applied frequently for many small animals, which provided accurate estimates of genome size in many previous reports (such as Liu K. et al., 2017, GigaScience, 6(4): giw012). See more details in lines 77-83.

L122: polish -> polishing Reply: Sorry for the typo. It was corrected in line 130 of the revised manuscript.

L123: I would have expected to see some information concerning Hi-C library preparation and sequencing, which are not present here. This is an important point for a data note article. Reply: Thanks for your instructive advice. Yes, it is done. More details regarding the Hi-C library preparation were provided in lines 110-113 and 131-142.

L152: "Ostreae Concha, Fucata martensii". There is an error here, as these are not correct scientific names. The authors probably refer to Pinctada fucata martnesii, and I don't know what Ostreae Concha is. If this is not a species, then the species count would be eight, not nine. Also penaeus should be Litopenaeus. Reply: Sorry for the mistakes. We revised Ostreae Concha (oyster) as Crassostrea gigas in lines 164. Meanwhile, Pinctada fucata martensii was used to replace Fucata martensii, and Litopenaeus vannamei was also provided in lines 164-165 of the revised manuscript.

L164: "by aligning against". Please be more specific, as this is also a BLAST-based analysis. Reply: Yes, we replaced it with “a BLAST-based analysis” in line 176.

L166: again, this is not an alignment, but rather a detection of HMM models. Reply: Yes, it is done. See more details in lines 195-197.

L185: several details are missing here, including the molecular model of evolution set (and a description of how it has been selected). Reply: Thanks for your nice advice. Yes, we added more details as follows in lines 195-197. Alignments of these ‘supergenes’ were carried out to construct a phylogenetic tree by using the Maximum Likelihood method in PhyML (v3.0, RRID:SCR_014629) with the HKY85 model and default parameters [36].

L209: DEGs: which comparisons were investigated here? I guess between the two seasons, but please be more specific here. Reply: Yes, you are right. We made modifications as follows in lines 220-223. The Cuffdiff in the Cufflink package with parameters of “-FDR 0.05 – geometric-norm TRUE –c 10” was utilized to predict differentially expressed genes (DEGs) in the testis and androgenic gland between reproductive season and non-reproductive season.

L210: folds should be "fold change" Reply: Yes, it is done in line 224 of the revised manuscript.

L221: missed -> missing Reply: Yes, it is done (line 235).

L248: "four iag genes" -"four paralogous iag genes"? Reply: Yes, it is done in line 262.

L249: rather than the number of genes, it would be much more interesting here to get to know the distance in MB between the 3 genes Reply: Thanks for your good advice. Yes, the distance covering these three iag genes was calculated to be 17.34 Mb. This sentence was therefore revised as follows in line 263-264. The distance covering the three iag genes was 17.34 Mb with prediction of 363 genes in this area.

L291: "lower organism"??? Reply: Sorry for the misleading description of "lower organism". We deleted this sentence in the revised manuscript.

Discussion: I found it very disappoint to find no mention at all about the predicted WGD event. This interesting finding is only mentioned in the results section, and not discussed at all. Was this expected? Are other WGD events known in Palaemonidae or Pleocyemata? Reply: Sorry for this missing discussion for the WGD event. We added a paragraph of WGD in the discussion section (lines 295-300).

How do the number of assembled choromosomes and genome size compare with cytogenetic estimates? I see, from the animal genome size database, various c-value estimates from different Macrobrachium species, ranging from 6.48 to 22.16 (but data from M. nipponense is not available). Reply: Thanks for your good question. In fact, there is no report of predicted genome size of M. nipponense from a flow cytometry experiment. We here estimated its genome size is about 4.6 Gb. However, from those genomic estimates in our previous genome works, the predicted genome sizes from a kmer-analysis are usually similar to the data from flow cytometry experiments. Therefore, we are confident about the estimate, although corresponding cytogenetic estimate is expected to be done whenever it is possible.

A number of studies exist concerning Macrobrachium karyotype. According to Damrongphol and colleagues (1991), the aploid number of chromosomes in M. rosenbergii is 59, and this value is confirmed in M. malcompsonii (Rebecca et al. 2020). M. lachesteri has n = 58 (Phimphan et al. 2018). Others, like M. acanthurus and M. amazonicum have n=49 (Molina et al. 2020), as reported here, but this should be mentioned in the manuscript. However, I think a key point that would be worth mentioning is that the number of assembled chromosomes matches with the expected number from karyotype studies. Reply: According to your advice, we cited the studies of Macrobrachium karyotype in lines 287-294.

More importantly, is there any clear evidence that Macrobrachium spp. has undergone a WGD event? I think this should be supported by much more substantiated evidence, which would allow a more precise timing for this event. Reply: No. However, we provided few consistent evidences for the WGD event in our manuscript. At first, in Fig. 2b (pink lines), we found large numbers of synteny blocks that were evenly distributed in each chromosome. On the other hand, we predicted 4dtv values from the gene synteny blocks. We observed a remarkable peak for M. nipponense in Fig. 3b, which predicted that a recent WGD event had occurred in the M. nipponense genome. Subsequently, we reconstructed that the 4dtv peak (blue) for L. vannamei was located at 0.94 (in the X-axis), which is similar to previous reports; the 4dtv peak (red) of M. nipponense was at 0.335. Meanwhile, we predicted the divergence between M. nipponense and L. vannamei was about 327.5 million years ago. Therefore, the WGD event in M. nipponense may appear about 109.8 million years ago.

Honestly, the closest genomes included in this analysis were those of L. vannamei and P. virginalis, but these species are Pleocyemata and quite distantly related with M. nipponense. Why were not the genomes of Pandalus platyceros, Caridina multidentata and Palaemon carinicauda used, since these are evolutionarily closer and available? Reply: Thanks for your good questions and instructive comments. We quite agree with you to use more crustacean species for a better phylogenetic tree. We considered that a chromosome-level genome assembly with continuous scaffolds would be more appropriate for a comparative phylogenomic analysis. However, those key species you recommended are reported with low-quality assemblies. For example, Pandalus phatyceros (scaffold N50: 1,512 bp; NCBI accession number: GCA_005815305.1), Caridina multidentata (scaffold N50: 819 bp; GCA_002091895.1), and Palaemon carinicauda (scaffold N50: 962 bp; GCA_004011675.1/) are too fragmented to be used for this phylogenetic analysis.

In summary, the WGD point should be revised and probably just mentioned as a possibility suggested by the analyses carried out, but whose timing needs to be properly investigated in future works. Reply: Thanks for your advice. Yes, we added a brief discussion of the WGD in lines 295-300.

Table 2 is incomplete. What do these numbers represent? Fold change values? Reply: You are right, the numbers in Table 2 represent fold changes. Related description was provided in Table 2.

Reviewer #2: This genome note describes the chromosomal assembly of the Oriental river prawn. It appears to be well assembled into chromosomes with a high BUSCO score. There gene models are reasonable given the whole genome duplication but someone in the future may wish to redo the annotation with the more recent version of Maker (version3) as they use version 2 which does not take advantage of EVM. This is a significant advance in the genomic resources for this clade of organisms and I recommend that it be published with some minor revisions. 1) The use of the term "lower organism" should be avoided as it is an outdated term to imply there are move and less evolved organism in the tree of life when in fact all organisms are equally evolved. Please revise the following sentence. "The Oriental river prawn is a lower organism; thus, it makes sense that sex-related genes 292 were not enriched in a special location (Figure 4)." Reply: Thanks for your nice advice. This sentence was deleted in the revised manuscript.

2) There is a lot of significance place on this one region on chromosome 25 where 3 iag genes are located in which some of the gene models between these three genes are also differentially expressed between the sexes. Based on the evidence, I feel this language should be toned down in their discussion and conclusions. For example, this sentence. "Thus, "Signal transduction and Endocrine system metabolic pathways", and the DEGs in these two metabolic pathways, might dramatically affect the process of male sex-differentiation and development in the Oriental river prawn." Reply: Yes, we agree with you. According to the comments from Reviewer 1, we limited descriptions of DEGs to a bare minimum. This sentence was revised as follows in lines 307-311. A few plausible sex-related candidate genes were identified, particularly after combining the analysis of genes on Chromosome 25 and differential transcription in testis and androgenic gland between the non-reproductive and reproductive seasons. However, these results require more independent validations. Meanwhile, similar correction was realized in the Conclusion section (lines 319-322).

The paragraph above this sentence defines the functions of each gene and then proclaim a final sentence saying Thus it follows. Honestly, it is not terribly clear from the definitions of the DEG genes in the above paragraph how it justifies a Thus statement. Now that they have a genome they could resequenced or perform GBS on 50 males and 50 females and determine if there is a sex locus and use that as stronger evidence for linkage of these genes to the sex determining region and then explore genes in that region that may be involved in the observed sexual body size dimorphism. Now the authors have already performed significant work here and this would be above and beyond the current work. My point being that they have identified some genes of potential interest that could be followed up with additional experiments and that the language could be toned down a little to accent this point and not draw away the significant work contained within. Reply: Thanks for your instructive advice. We have limited the descriptions of DEGs to a bare minimum, and toned down the conclusions in lines 307-311. In this data note, we just provided some plausible sex-related gene candidates. In our further plan, we will determine the detailed functions of these genes through RACE cloning, qRT-PCR analysis, in situ hybridization, and CRISPER-Cas9 knock-out; meanwhile, we are planning to identify some sex-related loci in these genes.

3) I can't find JACEGS000000000 in genbank at NCBI. Please be sure to release this data. Reply: Yes, the chromosome-level genome assembly is accessible now. Please try it once more for public availability.

Thank you for providing a revised version of your MS along with detailed replies to my comments. I find that most of my concerns have been adequately solved. However, a few additional modifications would be needed in my opinion.

I understand the rationale for using multiple individuals, and I appreciate the strategy of selecting multiple individuals originated from the same parents to minimize the issues linked with heterozygosity in the assembly. I have, however, a strong objection here, as k-mer-based estimates of genome size are not reliable under such circumstances, due the possible non-perfect overlap of the peaks generated by gDNA from multiple individuals, as the quantity of gDNA present in the pool from each individual are unlikely to be identical. I therefore strongly suggest the authors to remove the mention of k-mer based genome size estimates, as these would be only reliable with the analysis of Illumina reads generated from a single individual.

Line 197: how was the HKY85 model selected? Did the authors use ModelTest to identify the best-fitting model of molecular evolution?

Figure 3a: please indicate what do the numbers indicated between brackets mean (confidence intervals, I guess)

Lines 298-300: please improve the English language used here.

The WGD discussion remains unconvincing. The authors mentioned Yuan et al. 2017 (the E. carinicauda genome) stating that this genome was not characterized by a WGD event. Actually, both genome size (5.7Gb) and 2n chromosome number (90) do not indicate significant differences compared with M. rosembergii. On the other hand, the number of genes is more than 2X higher in E. carinicauda. The quality of the E. carinicauda genome assembly was much lower than the genome reported in this study, so do the authors think that clues linked with WGD might have been missed in the study by Yuan et al.? Please also note that other Palaemonidae species have been previously shown to be characterized by very large genomes. For example Palemon serratus, with a c-value > 10. Overall, I believe that the timing of this WGD event cannot be defined with certainty at the present time due to the lack of genome and karyotype information for many species, so I would appreciate if the authors could state more explicitly that many uncertainties about this inference remain.

Declaration of competing interests Please complete a declaration of competing interests, considering the following questions: Have you in the past five years received reimbursements, fees, funding, or salary from an organisation that may in any way gain or lose financially from the publication of this manuscript, either now or in the future? Do you hold any stocks or shares in an organisation that may in any way gain or lose financially from the publication of this manuscript, either now or in the future? Do you hold or are you currently applying for any patents relating to the content of the manuscript? Have you received reimbursements, fees, funding, or salary from an organization that holds or has applied for patents relating to the content of the manuscript? Do you have any other financial competing interests? Do you have any non-financial competing interests in relation to this paper? If you can answer no to all of the above, write 'I declare that I have no competing interests' below. If your reply is yes to any, please give details below.

I declare that I have no competing interests.

I agree to the open peer review policy of the journal. I understand that my name will be included on my report to the authors and, if the manuscript is accepted for publication, my named report including any attachments I upload will be posted on the website along with the authors' responses. I agree for my report to be made available under an Open Access Creative Commons CC-BY license (http://creativecommons.org/licenses/by/4.0/). I understand that any comments which I do not wish to be included in my named report can be included as confidential comments to the editors, which will not be published. I agree to the open peer review policy of the journal.

Authors' response to reviews:

Reviewer #1: Thank you for providing a revised version of your MS along with detailed replies to my comments. I find that most of my concerns have been adequately solved. However, a few additional modifications would be needed in my opinion. I understand the rationale for using multiple individuals, and I appreciate the strategy of selecting multiple individuals originated from the same parents to minimize the issues linked with heterozygosity in the assembly. I have, however, a strong objection here, as k-mer-based estimates of genome size are not reliable under such circumstances, due the possible non-perfect overlap of the peaks generated by gDNA from multiple individuals, as the quantity of gDNA present in the pool from each individual are unlikely to be identical. I therefore strongly suggest the authors to remove the mention of k-mer based genome size estimates, as these would be only reliable with the analysis of Illumina reads generated from a single individual. Reply: Thanks for your instructive comments. Sorry for the misleading descriptions in the previous version. In fact, the Illumina reads for our k-mer analysis was generated from a single individual. We extracted gDNAs from the muscle of this individual, and constructed two short libraries with the insert sizes of 500 and 800 bp respectively for the Illumina sequencing. The whole genome size was then estimated to be 4.6 Gb by the k-mer analysis. However, we thought the gDNAs from a single individual would be insufficient for whole genome sequencing. Therefore, we collected five more individuals for Illumina sequencing and another batch of five individuals for Pacbio Sequencing, in order to provide sufficient gDNAs for the practical whole genome sequencing. Please find more details in the revised manuscript (lines 80-93).

Line 197: how was the HKY85 model selected? Did the authors use ModelTest to identify the best-fitting model of molecular evolution? Reply: In our study, PhyML was employed to construct the phylogenetic tree with the default model HKY85. Thus, we tried to identify the best-fitting model to revise the phylogenetic tree by using the jModelTest. The best-fitting model was identified as GTR. Based on the GTR model, we rebuilt the phylogenetic tree, (Drosophila_melanogaster,(Daphnia_pulex,((Pinctada_fucata,Ciona_savignyi),(Macrobrachium_nipponense,(Procambarus_virginalis,Litopenaeus_vannamei))))). However, this revised tree has the same topology as the previous one. After fossil calibration of divergence times with the MCMCTREE model in PAML, we also observed that the revised tree has no difference from the previous one.

Figure 3a: please indicate what do the numbers indicated between brackets mean (confidence intervals, I guess) Reply: Yes, you are right. The numbers within brackets represent confidence intervals. Thais sentence was added to the figure legend of Figure 3 (line 386).

Lines 298-300: please improve the English language used here. Reply: Thanks for your advice. Yes, it is done in lines 304-307 of the manuscript.

The WGD discussion remains unconvincing. The authors mentioned Yuan et al. 2017 (the E. carinicauda genome) stating that this genome was not characterized by a WGD event. Actually, both genome size (5.7Gb) and 2n chromosome number (90) do not indicate significant differences compared with M. rosembergii. On the other hand, the number of genes is more than 2X higher in E. carinicauda. The quality of the E. carinicauda genome assembly was much lower than the genome reported in this study, so do the authors think that clues linked with WGD might have been missed in the study by Yuan et al.? Please also note that other Palaemonidae species have been previously shown to be characterized by very large genomes. For example Palemon serratus, with a c-value > 10. Overall, I believe that the timing of this WGD event cannot be defined with certainty at the present time due to the lack of genome and karyotype information for many species, so I would appreciate if the authors could state more explicitly that many uncertainties about this inference remain. Reply: Thanks for your comments. Yes, you are right; we really think the important clues of WGD event have been missing in the Yuan’s study, because the contig N50 of the E. carinicauda genome is only 263bp. These data are indeed fragmental to be used for gene identification. Therefore, too many genes and synteny blocks in the E. carinicauda genome were missed. For the genome size, rich repeat content could be a major reason for the genome expansion. Based on our experience, the Palemon serratus could have much more repeat content. For the WGD analysis, we employed both 4dtv and synteny block analyses that were well-performed and classical methods for WGD identification in various animal and plant genome reports. In the coming future, if there is any newly published genome assembly of a close shrimp species, we will further analyze related WGD event/time. According to your advice, we rewrote related sentences. Please find more details in lines 301-307.