Overview The paper describes two genome assemblies of penguins, their annotation and selection tests in four groups of candidate genes believed to have significant changes in penguin evolution.

The distinctive penguins adaptations allow them to live an aquatic life in a variety of environments, from the hostile environments in Antarctica to the equatorial Galapagos. This paper focuses on two penguin species Adélie penguin (Pygoscelis adeliae) and emperor penguin (Aptenodytes forsteri).

The current study provides new data for the interpretation of the speciation, changes in effective population sizes over the past 10 million years, and the putative associations between the whole-genome patterns and climate changes. In addition, it focuses on molecular changes in four groups of genes related to

1. epidermal structure,

2. photo-transduction,

3. lipid metabolism, and

4. forelimb morphology.

Overall, while this paper is timely and worthy of publication but it suffers from the simple fact of being a companion paper to a much larger manuscript to which I have no access at this time. Therefore it was difficult to address some of very interesting issues. I was hoping that he paper is written from top down – from the larger prospective of genome evolution in birds. However, evolution of penguins is a fascinating topic, and I have to admit, I have learned a lot while reading the background papers preparing to this review. Therefore, I would hope that the authors would treat it more with a consideration of an independent study, given the unique and important contribution it is making to the study of vertebrate evolution.

The manuscript is given in the form of a research paper, but resembles more of the data communication. This is partially explained by the companion status of this manuscript. This is why in addition to the comments on the annotation; I have outlined a few recommendations that could improve the data interpretation.

Revisions and Comments

• Sequencing, Assembly, and Annotation Comments The small contigs that average 19-30 kb in the two species joined by an array ofmate pairs from 2, 5, 10 and 20kbp have resulted in the 5 Mb scaffolds suggestinga great quality assembly as authors indicate – comparable to that of zebra finch and turkey, however it is not clear how good these scaffolds which seems are and how many gaps and chimeric sequences are expected. This may be leading to a more optimistic estimate of genome coverage.

• Gap width estimates: Unsequenced regions between mate pairs in contigs and between scaffolds are often represented as runs of ‘N’s in the final assembly. Thus two assemblies can have identical scaffold N50s but can still differ in their percent gaps: one has very few gaps, and the other is heavily peppered with them. This can be addressed by providing the gap width estimates.

• Are there expected chimeras? Are there possible chimeras in the assembly?

• Coverage and genome size estimates: The authors calculate the coverage, based on the contig length is 85%, but for some reason they compare it to their own k-mer based estimate instead of the C-value for penguins (~1.63pg).

• However, this is deceptive, as the contig coverage/C-value will be a more accurate prediction of genome sequence. I suggest that the authors use http://www.genomesize.com/search.php to look up the experimentally measured genome size for their calculations for the closest species (Gregory et al., 2007). d. What about the missing repetitive elements? Genes are easy to assemble, so that CEGMA estimate seems to be very good, but the repeat percentage (6-7%) seems to be small compared to what is expected for bird chromosomes (15-20% (Burt, 2002)), suggesting that a larger part of the repetitive genome is not assembled.

• Also, interesting is the fact that emperor penguin has three times the number of tandem repeats. Evolution of repetitive elements decline in birds has been contribute to the increased constraint (Burt, 2002), and flying birds have has less of the repetitive load then the flightless birds (Organ et al., 2007). Therefore, I expect the there are a lot more of the repetitive genome, that the current coverage might be suggesting.

Discussion comments

I think that the new data presented in the paper has a lot of potential to contribute to our understanding the causes of penguin evolution. Specifically, was the split between the two species a sympatric driven by the life history differentiation, or allopatric driven by the geographic separation? Was there a speciation by drift followed by selection in each lineage, or evolution was guided by natural selection from the start?

• I will be referring to Figure 1 in he future comments which be found in: Figure 1. Reconstruction of Earth's climate over the past 65 million years. Zachos, J.; Pagani, M.; Sloan, L.; Thomas, E.; Billups, K. (2001). "Trends, rhythms, and aberrations in global climate 65 Ma to present". Science 292 (5517): 686–693 http://www.sciencemag.org/content/292/5517/686 Molecular dating used by estimated that penguins diverged from their closestrelatives ~60.0 million which is more or less consistent with the previous estimates, but estimated divergence time between Adélie and emperor penguin is set at 23 My, almost double than that in Subramanian et al. (2013), and around the time the previous estimate of the crown penguin’s MRCA existence.

• This estimate has huge confidence intervals (6.9-42.8 Mya). A recent paper by Subramanian et al. (2013), calibrated for the fossil record, suggested that the age of the most recent common ancestor of extant penguins to be 20.4 Mya (17.0–23.8 Mya), and most of the major groups of extant penguins diverged 11–16 Mya. This, as the authors argue, overlaps with the sharp decline in Antarctic temperatures that began approximately 12 Mya, suggesting a possible relationship between climate change and penguin evolution.

• The current paper argues for a much earlier date, and disagrees on the exact timing of the penguin lineages, is based on molecular evidence alone (no fossil calibration). I hope that a stronger argument is made for the different date, because coupled with the climate data, this difference has implications to interpretation of the causes of penguin speciation. Was speciation in the two penguins caused by the adaptation to the cold aquatic environment? The answer is probably a “yes”, but does this paper clearly show what it claims in the title?

• It is not clear that adaptations addressed in the paper are directly caused by adaptation (selection): There is definitely some evidence pointing to the similar adaptations in penguin species, such as expansion of keratinacyte beta keratins: 13 and 15 genes in penguins (Adele and Emperor) vs. 6 and 7 in the closest related species (egret and ibis).

• Not conclusive is the adaptation argument for evoplankin and DSG1 - Thick soles are important to walk on ice, but it is clearly an adaptation for cold climate, rather than walking lifestyle like it is in humans? There can be several alternative explanations to the question how did the penguins develop flightlessnes, or as it is addressed here “changes in forelimb structure”. At the time of the split between penguins and albatrosses the Antarctic climate was much warmer.

• Abundant fossil leaves and wood point to the existence of forestation even a few degrees of the South Pole. Marsupial mammals once lived where there is now ice – a fact supported by a few fossils so far, but reasonable to assume. Extinction of terrestrial mammals in the glaciation period would allow birds in Antarctica to become flightless, just like that in New Zealand. Similar expectations would be in place if penguins evolved by adaptation to reproduce on new drifting ice during glaciation.

• Was this adaptation (selection) or relaxed constraint (drift)? It is no surprise then that no genes were detected with selection signatures in the forelimb structure, while some accumulated random non-neutral substitutions – the evolution of flightiness was first likely a relaxation of selection for flight. Some of these are might be older than others – the EVC2 seems to have the same mutation in both penguins indicating that the flightnessness may have occurredin penguins before they have separated from each other. Other reported changes, such as the loss of the middle sensitivity Rh2 opsin gene is hardly an adaptation, and can be attributed to the relaxed constraint.

• Remarkably it is only the middle (green) opsin lost, which leaves the range of the perceived light the same, since the short and long wavelength sensitive opsins (red and blue) remain. The loss of circadian gene, is hardly an adaptation – relaxed constraint is a better explanation, as day and night have a different meaning in Antarctica. Hardly selection-driven, these genes may have been affected by altered light/dark regimes, rather than cold aquatic environment. Finally, selection for different genes associated with light transduction - CNGB1, MYO3A and UACA - in emperor, and CRB1, CRY2 and MYO3B in Adelie, suggest that the adaptations were independent of each linege, and occurred after the species separated from each other, so they cannot be contributed to the common adaptation to cold aquatic environment, but to the species differences in reproductive strategy: one species reproduces in the dark night (Antarctic summer), the other one in the daytime (Antarctic winter).

• Differences between the two species: Interestingly, the major biological difference between the two species, the reproductive strategy, is only discussed in the chapter on effective population sizes. Emperor penguins do not require ice-free breeding grounds to incubate their eggs, and optimal body fuels during winter in order to succeed in their breeding, while in Adélie penguin clutch initiation and hatching dates depend on the temperatures. Both species need very large pieces of sea ice during molting so they don't drown, which may be why penguins have expanded with currents and diversified during the cold periods, providing an example of allopatric speciation. (Baker et al., 2006).

• Emperor penguins need to stay on fast ice or very large pieces of sea ice that remain stable for at least four weeks because during this time there feathers are not waterproof and they would sink and die if they had to swim. While the genome pair may have served as a poor choice to find adaptations for cold aquatic environment, the two genomes will become a great resources for discovering candidate genes that could be involved in these differences.

• Time of divergence is not an arbitrary number. The one issue with the two dates is that on one side (~35-25 Mya) is Antarctic glaciation; while on the other side (~25-12 Mya) is Antarctic thawing followed by re-glaciation (~12-14 Mya)(see Figure 1 below). So what is driving penguin speciation, warming or cooling? Since there is a common ancestor to the two penguins who became flightless and survived on Oligocene ice, it seems that the timing postulated in this paper places species divergence at the time of the first Antarctic warming - thaw. Subramanian et al. (2013) estimates for most of the major groups of extant penguins were to diverge 11–16 Mya, would put diversification of penguin taxa in the cooling period resulting in the formation of large ice sheets in Antarctica (Baker et al., 2006) (see Figure 1). e. Species choice – may have helped in this discussion. In the times of cooling, penguin species reacted differently. Some moved out to the warmer places as itwoud give them better places to reproduce. In fact, it is only the representatives of Artenodytes and Pygocelis who stayed and adapted to the cooler conditions. Thus to address the adaptation to cold, would be desirable therefore to choose a penguin species from the warmer latitude for the reference in search for the cold-adaptation genes. I realize that authors probably did not have a choice in the matter.

• For the above reasons, I disagree with the manuscript title: “The genomes of two Antarctic penguins reveal adaptations to the cold aquatic environment” because of emphasizing adaptation, which would imply causal link to the unequivocal signatures of natural selection. I think the more appropriate title will be “The genomes of two Antarctic penguins have been shaped by millions of years of life on Antarctic ice”. Incidentally, mentioning disappearing “ice” in the title can help tie the topic to the recent warming trends threatening survival of both of these majestic species.

Minor

• Sentence and word omissions, for example - Genome annotation section (p10) is missing something at the end of the paragraph the sentence abruptly ends

• Consistency in reporting data, for example, NC RNA has number for miRNA, I would be nice to list numbers for ncRNA and tRNA in the text as well

• Numbers below 10 should be given in words. For example, P 12 first paragraph – “6 closely related aquatic species”, should be spelled out as “six closely related aquatic species”.

References

• Baker AJ, Pereira SL, Haddrath OP, Edge KA (2006) Multiple gene evidence for expansion of extant penguins out of Antarctica due to global cooling. Proceeding of the Royal Society of London Series B (Biological Sciences) 217: 11–17.

• Burt DW (2002) Origin and evolution of avian microchromosomes. Cytogenet Genome Res 96: 97–112.

• Subramanian S, Beans-Picón G, Swaminathan, SK, Millar CD Lambert DM (2013) Evidence for a recent origin of penguins. Biology Letters 9, 20130748.

• Organ, C L., Shedlock, A.M., Meade, A., Pagel, M. & S.V. Edwards (2002) Origin of avian genome size and structure in non-avian dinosaurs. Nature 446, 180-184

• Gregory, T.R., Nicol, J.A., Tamm, H., Kullman, B., Kullman, K., Leitch, I.J., Murray, B., Kapraun, D., Greilhuber, J., and M. D. Bennett (2007) Eukaryotic genome size databases. Nucl. Acids Res. 35 (suppl 1): D332-D338

Level of interest:

An article of outstanding merit and interest in its field

Quality of written English:

AcceptableStatistical review:No, the manuscript does not need to be seen by a statistician.

Declaration of competing interests:

I declare that I have no competing interests