Talk:PLOS/Eukaryote hybrid genomes

Reviewer 1: Alexandre Blanckaert
ID version: 7469

Summary The authors present here a detailed overview of hybridization and its consequences. They described its various outcomes focusing on the cases where the final genomes present genetic content from both parent species: adaptive introgression and hybrid speciation. The consequences of hybridization in terms of genetic barriers are presented and discussed. The story is nicely illustrated by many biological examples.

Major remark - the authors need to make a clearer distinction between prezygotic versus postzygotic incompatibilities and intrinsic versus extrinsic ones. For example, in the abstract the authors wrote : “Homoploid hybrids can sort the intrinsic barriers that isolate the parent species from each other so that a subset of these isolate the hybrid species against each parent species, or gain pre-mating extrinsic isolation from ecological differences.” The term postzygotic is here directly associated to intrinsic and prezygotic to extrinsic. This is misleading as any combination can exist in reality. (see Coyne and Orr 2004 table 1.2 p29, Turelli Barton Coyne, 2001 TREE). This is a problem that is somewhat pervasive in the whole manuscript, see also for example the definition of postzygotic incompatibilities in the glossary. I would suggest that the authors clarify the respective definitions and revise the whole manuscript accordingly.

Minor remarks

General - I would suggest to also mention autopolyploidisation. Although this not technically hybrid speciation, it shares many properties with allopolyploidization.

Abstract - genome stabilization: it is not so much about the fixation of compatible combinations of genes than the loss of the incompatible ones. For example in an aAbB system, where A and B are incompatible, the loss of (let’s say) A means that the incompatibility is gone, however, it does not imply fixation at the remaining B locus. It could be good to clarify this in the text. - hybrids do not have to be unfit to form a stable hybrid zone - low population density or local adaptation is also sufficient to explain the stability of a hybrid zone [Barton 1979] - in the abstract “incompatible combinations of genes derived from” is a bit weird - I would recommend to rephrase to "incompatible combinations of alleles originating from" as most models of incompatibilities focus on interactions of alleles (BDMI).

Evolutionary outcomes of speciation - “reduction of fitness of parents” - I would recommend to mention that this reduction in fitness increases the risk of extinction of the parent populations. This addition in the text will fit together with the addition of extinction as an outcome in Figure 1.

Adaptive introgression - l2 there is a typo in "species" - I would recommend to cite some theoretical literature on adaptive introgression here - for example Barton and Bengtsson 86, Demon 07, Uecker 15

Homoploid and polyploid hybrid speciation - I would recommend to add that it is necessary that the genome of the homoploid hybrid contains tracks of the genome of the two parental - otherwise it will fail to establish reproductive isolation. It requires one of the following: novel phenotype matching strong exintisic pressure to avoid niche overlap, different premating preferences, or reciprocal sorting of the various incompatibilities to establish postzygotic barriers with both parent species. - I find the last sentence a bit confusing as the second part refers back to homoploid speciation, whereas the former 3 were about polyploidisation

Reproductive isolation against parent species - I would recommend to define phenology here or add it to the glossary. - The authors use the term “transgressive traits", then "transgressive phenotypes" but define "transgressive segregation" in the glossary. It would be easier for the reader if the same terminology was used (trait/phenotypes or segregation).

Hybrid genomes and genome stabilization - This section is very lengthy. I would suggest to split it into 2 parts depending on the outcome. For example, the first three paragraphs in this section could be kept, and a second section called “Genome-wide consequences of hybridization” (or something similar) could be created, corresponding to the last 4 paragraphs there. - “recombination is suppressed near genes” - I find this somewhat misleading as there is no mechanistic reduction of recombination - what ref 51 shows is that the ancestry blocks are on average longer. Recombination events are not rarer - this reduction in effective recombination rate stems from individuals not recombining having on average a higher fitness. I would recommend to probably rephrase to “effective recombination rate is reduced, due to the lower fitness of the recombinants, more likely to carry incompatible alleles along the same chromosome”. This sentence confused me because at first I understood that there was selection for a reduction in recombination rate (modifier model) around DMIs. - “the time to genome stabilization is variable”. In a simple model, Blanckaert & Bank 2018 showed that this time is highly influenced by population size. - “the balance between allelic … what introgression tract will be inherited”. The relative frequency of the parental genotype is a major force here (because of frequency dependent selection) Genetic drift as well as the linkage architecture also plays a role. This is especially true for hybrid species. I would recommend to add these two effects to the paragraph for completeness.

Conclusions - “As hybridization is frequent, ...” I do not understand the causality link with the next part of the sentence. Also “hybridization is frequent” has two meanings - either this phenomenon happens for a lot of species pair or lots of hybrid individuals were produced from the hybridization of a given pair. It would be great if the authors could clarify this. - First sentence of the second paragraph: I would recommend to change to “....evidence that … is common, contributes novel species and ...”

Figure 1: One outcome of hybridisation in Fig 1 is missing, namely successful invasion of one species by the other and I would recommend to add it to the figure. Although invasion can happen without hybridization, loss of one species is still a possible outcome, as hybridization creates a fitness cost on both parental species (DMI - but also search cost if assortative mating).

Figure 2: Homoploid speciation and F1 hybrid are written on top of each other Choice of the polyploid illustration is slightly confusing - as having the F1 hybrid between parental species infertile is not the common case.

Figure 3: Unless I am mistaken, the stack of haplotypes here represents a population and not a genotype (as in Figure 2). I would recommend to use a different representation to make it more clear. I was wondering whether the authors intend to keep this file (nomination for deletion from one of the authors).

Glossary - hybrid genome: form -> from - postzygotic incompatibilities: The authors shortly define first postzygotic incompatibilities as intrinsic postmating barriers and then the definition is expanded to include extrinsic barriers as well. I would recommend to replace the first sentence by: “genetic barrier due to the offspring of a mating being unfit or infertile”.

Reviewer 2: Gonzalo Nieto Feliner
Summarizing all the relevant aspects of a complex and wide-ranging topic as hybridization is not easy, particularly in what refers to its creative role in evolution. In my opinion, the article is balanced, covers all relevant topics and is thus useful. I only have a few suggestions. While it is classic to distinguish adaptive introgression and homoploid hybrid speciation as different outcomes of hybridization, based on some of the studies on Heliconius butterflies I suggest that the possibility that introgressed lineages acquire reproductive isolation and give rise to a hybrid species is considered in the ms.

“Evolutionary outcomes of hybridization”: Reinforcement could be mentioned as one possible outcome of hybrids that are not viable or sterile. For instance, Phlox drummondii (Hopkins, New Phytol. 2013). Under this same heading or under the following one (“adaptive introgression”), perhaps you could mention the possibility of neutral (non-adaptive) introgression.

“for most suggested examples of homoploid hybrid speciation, the genetic basis of reproductive isolation is still unknown”. There are others in which external isolation appears to have done the job (e.g., Senecio squalidus, in which hybrids were brought to the UK by the beginning of the 18th century away from their progenitors)

“Hybrid species can occupy an ecological niche different to those of the parents and may be isolated from the parent species primarily through pre-mating barriers (introgressive speciation, c.f. [30])”. I am not aware of the use of such term (introgressive speciation) in that sense. I assume you refer to what Grant called “hybrid speciation with external barriers”. Incidentally reference #30 appears as a mix of two: the classic Verne Grant’s Plant Speciation (1981) and a paper by Payseur and Rieseberg (2016) A genomic perspective on hybridization and speciation.

“A similar pattern was found in Galapagos finches where a specific hybrid beak shape and song are thought to have been important [8]”. In a section on reproductive isolation I suggest removing the reference to the beak shape since for reproductive isolation it’s song what is relevant. “In what taxa is hybridization common?” Maybe other potential differences between plants and animals that could underlie the different % of hybrids estimated by Mallet in the two groups could be mentioned. E.g. reproductive isolation (Widmer et al., 2009, Heredity)

“Paradoxically, the ability to self-pollinate may also act in favour of stabilising allopolyploid taxa”. This is also the case in recombinational speciation. “Recombinatorial hybrid speciation: the process by which a hybrid taxon develops reproductive isolation from the parent species through sorting of incompatibilities”. Here and elsewhere I suppose you want to say “Recombinational” (cf. Grant, 1971). In addition, I suggest saying “a hybrid population” instead of “a hybrid taxon” since the population(s) developing reproductive isolation need not yet be considered a taxon.

There are also a number of typos across the text as well as some inconsistent use of UK and US English spelling and some missing words (e.g., in the Box on genomic tools, “Other phylogenetic network methods that [something missing here] incomplete lineage sorting and hybridization may also help [131,132].”

Reviewer 3: James Mallet
I thought this was a very full and thorough review. I enjoyed reading it. I'd probably have written it somewhat differently myself in places, but it is very useful just the way it is.

I question the worth of doing this as a Wiki page, however. If I were Runemark et al., I think I would want this to be a proper published review article rather than a Wiki page. I use Wikipedia a lot, and indeed all of us probably do as a first port of call to find out quickly about stuff. But the point of Wikipedia is (a) that it is UNauthored; in other words, anyone can weigh in on it and change stuff they don't like, and (b) it has to be fairly brief for a different more public market than a scientific journal review article. Also, a review article should be a statement of the understanding at the time it was written. As knowledge advances, the review article will go out of date, and it's probably best to write a new review article rather than restructure the old one.

So, I would have thought that this half-way Wiki/review article is not going to be so useful... A Wikipedia page is sufficiently flexible that it can change. But of course, I could be wrong!

I've taken advantage of the Wiki format to make a few minor changes to grammar and typos directly on the document that I think the authors will approve. A few extra points and queries I will put here:

1) "In addition to recombination rate, the coupling ..." I do not understand what is meant here. Coupling is a strange term; I suggest it can be dropped for a more explanatory term. Can you expand and explain?

2) "Minor parent," "minority parent ancestry," etc. In general, we should always entertain the possibility that the donor of introgression may become the majority parent. Such things appear to have happened in the Anopheles gambiae group, and in several unpublished cases in Heliconius. I would prefer if you could use donor and recipient, although in some of the cases you discuss (Xiphophorus and humans) the donor is indeed the minority parent.

3) "asexual fusion of cells or hyphae". If it's a hybrid, surely it is sex!  I have edited the page to read "ameiotic fusion ..." -- I hope that's ok.

4) "Primary" and "secondary hybridization:" I really don't like those terms, though I know they have been used in the literature. It's reminiscent of primary and secondary contact, but surely it just refers to the age of the species before hybridization, rather than the geography involved (even though geography may sometimes of course contribute to a period of lack of hybridization).

Reviewer 4: Roger Butlin
“Hybridization is the process where genetically closely related species mate and produce offspring with admixed genomes.”

This is not the most commonly-used definition of hybridization. The term is more often used to describe mating and offspring production between genetically divergent populations. This can include interspecific hybridization if the species concept in use allows for the production of viable and fertile offspring. To restrict the term to crosses between species excludes many interesting cases, especially examples of hybrid zones.

“If hybrids are not inviable or sterile …”

‘Hybrids’ come in many different flavours! First generation hybrids have one copy of each genome (in a typical diploid and excluding sex chromosomes, which might cause important differences between the sexes). For weakly-diverged parents, they may show heterosis and they are unlikely to show incompatibilities. Later generation hybrids have admixed genomes with many genotypes possible and a wide range of possible fitness effects. Therefore, it is dangerous to refer to ‘hybrids’ as a single group without being more specific. In particular, ‘hybrid fitness’ always needs to be qualified to describe the class of hybrids and to consider variation in fitness among genotypes.

“Below we describe the evolutionary outcomes of hybridisation that result in hybrid genomes.”

I think the authors have in mind ‘persistent hybrid genomes’, where one or a small number of possible combinations of parental haplotypes comes to dominate a population. This is different from a hybrid zone, for example, where ‘hybrid genomes’ containing many different mixtures of parental alleles are continuously generated but also continuously removed by selection.

“Some researchers argue that evidence of a hybridization-derived basis for reproductive isolation should be an additional defining criterion of a hybrid species [26], but see [27].”

This debate is about ‘hybrid speciation’ rather than about ‘hybrid species’. Schumer et al. [26] would accept that a species with a genome made up of a stable mix of two parental genomes is a hybrid species, without evidence about the origin of reproductive isolation. However, to describe the process giving rise to this species as ‘hybrid speciation’ they would require evidence that reproductive isolation was generated, at least in part, by hybridisation.

“To establish successfully as a hybrid species, reproductive isolation against both parent species is required”

Reproductive isolation is a continuous quantity. Complete reproductive isolation is not required for successful establishment but it must be strong enough for the hybrid and parental populations to retain their distinct characteristics. “Recombination is more frequent in homoploid hybrid genomes than in allopolyploid hybrid genomes.”

Allopolyploids may have other ways to generate diversity and novelty, for example through differential loss or silencing of gene copies.

“Genome stabilization”

It is important to note that an isolated hybrid population will eventually stabilise for a pattern of homozygous blocks derived from the two parent species in the absence of selection. This can be taken as a kind of null expectation against which to compare observed patterns. In addition to the processes mentioned, many blocks may show heterosis because of the different sets of deleterious recessive alleles inherited from the parental taxa. This can act to slow down stabilization, especially in areas of low recombination.

“In addition to recombination rate, the coupling to incompatible loci may be important for the likelihood of retention of introgressed material”

The meaning of ‘coupling’ in this sentence is unclear.

“Schumer and Brandvain found that hybridization-derived regions on average are less constrained than the rest of the genome [54]. During the process of hybrid genome stabilization the minor parent genome content is overrepresented in high recombination areas, both in swordtail fish and in humans [31].”

What is meant by ‘less constrained’ here? Why are minor-parent haplotypes overrepresented in high recombination areas (presumably relative to low recombination areas, rather than to some external expectation)? Is it actually that the minor haplotype in underrepresented in low recombination areas because these areas retain incompatibilities that are expressed more often because the minor haplotype is more often in a ‘foreign’ background?

“In what taxa is hybridization common?”

This is quite a different question from, ‘In what taxa is formation of hybrid genomes common?’ On current evidence, hybrid zones are much more widespread than formation of stable hybrid genomes. The two outcomes may not be well correlated across taxa.

“Fungal hybridization may also result in asexual hybrid species”

This is also certainly true for animal speciation.

“Secondary hybridization is expected to have a greater potential to contribute beneficial alleles or generate novelty than primary hybridization because more divergent alleles are combined in secondary hybridization and are thus more likely to have a large fitness effect or to generate transgressive phenotypes”

Secondary hybridisation is also more likely to generate incompatibility (either intrinsic or due to maladaptation). Therefore, it is not clear that there is a simple monotonic relationship between divergence and the probability of generating persistent hybrid genomes.

“Hybrid genomes, including those of many individuals of our own species”

This seems to stretch the definition of a ‘hybrid genome’. The term loses its usefulness if it can describe an individual’s genome that happens to contain alleles derived from distinct ancestral populations. Earlier in thus article, it was used to describe stabilized genomes typical of populations. Unless there are Neanderthal or other haplotypes fixed for some parts of the human genome, the term should not be applied.

“Box: Detecting and studying hybridization with genomic tools”

It would be helpful to distinguish between detection of hybridisation and detection of persistent hybrid genomes.

Wikification
Please consider which terms could be usefully hyperlinked to wikipedia pages. You can provide these as a list or insert the links directly into the article using the following syntax: T Shafee (talk) 19:10, 10 May 2019 (PDT)
 * text to display
 * for example population structure is displayed as population structure.

Additionally, please upload the svg versions of figure 1 and 3 if possible. T Shafee (talk) 22:34, 22 July 2019 (PDT)


 * References now inline formatted with template. T Shafee (talk) 22:50, 27 July 2019 (PDT)

Response to referees
We have now revised the wiki-page in accordance with the insightful comments from all four referees. All comments have been fully considered, and in the vast majority of cases, we revised the text according to the suggestions of the referees. We think that including the perspectives suggested by the referees and revising the text in accordance with the referee’s comments has substantially improved this review, and we hope that this would now be suitable for publication as a topic page in Plos Genetics.

Some of the main changes we have done in the current version are:

1) Made a clearer distinction between intrinsic and extrinsic incompatibilities and discussed different forms of prezygotic and postzygotic isolation in light of this in response to Dr. Alexandre Blanckaert’s comments.

2) Discussed the possibility that introgressed lineages acquire reproductive isolation and reinforcement in response to Prof. Gonzalo Nieto Feliner’s comments

3) Rephrased and clarified “minority parent ancestry” in response to Prof. James Mallet’s comments.

4) Clarified definitions and revised our reasoning according to Prof. Roger Butlin’s suggestions.

Below, please find our detailed response to each of the reviewers’ comments.

Reviewer 1: Alexandre Blanckaert
Summary The authors present here a detailed overview of hybridization and its consequences. They described its various outcomes focusing on the cases where the final genomes present genetic content from both parent species: adaptive introgression and hybrid speciation. The consequences of hybridization in terms of genetic barriers are presented and discussed. The story is nicely illustrated by many biological examples.

1.1 Major remark - the authors need to make a clearer distinction between prezygotic versus postzygotic incompatibilities and intrinsic versus extrinsic ones. For example, in the abstract the authors wrote: “Homoploid hybrids can sort the intrinsic barriers that isolate the parent species from each other so that a subset of these isolate the hybrid species against each parent species, or gain pre-mating extrinsic isolation from ecological differences.” The term postzygotic is here directly associated to intrinsic and prezygotic to extrinsic. This is misleading as any combination can exist in reality. (see Coyne and Orr 2004 table 1.2 p29, Turelli Barton Coyne, 2001 TREE). This is a problem that is somewhat pervasive in the whole manuscript, see also for example the definition of postzygotic incompatibilities in the glossary. I would suggest that the authors clarify the respective definitions and revise the whole manuscript accordingly.


 * Response: We agree that this distinction is important, and have now made sure to properly distinguish the different types of reproductive isolation barriers. The revised section now reads “Reproductive isolation between a hybrid species and its parental species can arise from a variety of reproductive barriers either before or after fertilization (prezygotic or postzygotic, respectively), which may themselves be dependent or independent of environmental condictions (extrinsic or intrinsic barriers, respectively) {coyne2004speciation}. For example, intrinsic postzygotic barriers cause hybrid inviability or sterility regardless of the environment in which they occur, while extrinsic postzygotic barriers result in hybrids of low fitness due to maladaptation to specific environments [1]”. We have also made sure that other parts of the manuscript are written in a way that is consistent with this (separating pre-and postzygotic barriers and providing examples of both pre- and postzygotic of intrinsic and extrinsic barriers. We have updated the glossary and included these terms as well.

1.2 Minor remarks General - I would suggest to also mention autopolyploidisation. Although this not technically hybrid speciation, it shares many properties with allopolyploidization.


 * Response: We have added an explicit mention to autopolyploids. However, although an interesting possibility, since our review deals with interspecific hybridisation we have chosen to not discuss further the potential similarities that allopolyploids may have with intra-specific polyploids product of mating between genetically distinct populations. We added a sentence reading “In contrast to allopolyploids, autopolyploids are characterised by genome duplication within the same species and are thus not discussed further in the context of this review” to clarify this to the reader.

1.3 Abstract - genome stabilization: it is not so much about the fixation of compatible combinations of genes than the loss of the incompatible ones. For example in an aAbB system, where A and B are incompatible, the loss of (let’s say) A means that the incompatibility is gone, however, it does not imply fixation at the remaining B locus. It could be good to clarify this in the text. - hybrids do not have to be unfit to form a stable hybrid zone - low population density or local adaptation is also sufficient to explain the stability of a hybrid zone [Barton 1979] - in the abstract “incompatible combinations of genes derived from” is a bit weird - I would recommend to rephrase to "incompatible combinations of alleles originating from" as most models of incompatibilities focus on interactions of alleles (BDMI).


 * Response: We agree both that phrasing genome stabilization as selection against incompatible combinations is better and have now done so. We have also revised the sentence in the abstract according to the suggestion which now says “During the process of genome stabilization selection against incompatible combinations results in fixation of compatible ancestry block combinations within the hybrid species.”

1.4 Evolutionary outcomes of speciation - “reduction of fitness of parents” - I would recommend to mention that this reduction in fitness increases the risk of extinction of the parent populations. This addition in the text will fit together with the addition of extinction as an outcome in Figure 1.


 * Response: We agree, and have now incorporated the possibility of extinction into the figure legend “Finally, if hybridization leads to unfit offspring, it may reduce the fitness of the involved parental taxa due to wasted reproductive effort and may increase extinction risks for these.”. In addition, we mention this possibility in the main text “This could potentially lead to reinforcement, selection to strengthen premating isolation [2] or if the species fail to evolve premating isolation, it could increase their extinction risk due to wasted reproductive effort [3].”

1.5 Adaptive introgression - l2 there is a typo in "species" - I would recommend to cite some theoretical literature on adaptive introgression here - for example Barton and Bengtsson 86, Demon 07, Uecker 15


 * Response: Thanks for spotting the typo and providing interesting references. We have now incorporated a section on theoretical expectations for adaptive introgression and cited the suggested literature ”Simulations suggest that adaptive introgression is possible unless hybrid fitness is substantially reduced [4,5], or the adaptive loci are tightly linked to deleterious ones [6]”.

1.6 Homoploid and polyploid hybrid speciation - I would recommend to add that it is necessary that the genome of the homoploid hybrid contains tracks of the genome of the two parental - otherwise it will fail to establish reproductive isolation. It requires one of the following: novel phenotype matching strong exintisic pressure to avoid niche overlap, different premating preferences, or reciprocal sorting of the various incompatibilities to establish postzygotic barriers with both parent species. - I find the last sentence a bit confusing as the second part refers back to homoploid speciation, whereas the former 3 were about polyploidisation


 * Response: To make the text easier to follow we have divided the definitions, the description of the content of the stabilized genome and the reproductive isolation into different sections. We agree that to fully understand what a hybrid genome is all these components are needed, but we think we cover them well in the sections on “hybrid genome stabilization” and “reproductive isolation against parent species”. We have, however, reformulated the last sentence and moved it to the next section on reproductive isolation against parent species.

1.7 Reproductive isolation against parent species - I would recommend to define phenology here or add it to the glossary. - The authors use the term “transgressive traits", then "transgressive phenotypes" but define "transgressive segregation" in the glossary. It would be easier for the reader if the same terminology was used (trait/phenotypes or segregation).


 * Response: Thanks for bringing this to our attention. We have now defined “transgressive phenotype” in the glossary and used this wording throughout the text and added phenology to the glossary.

1.8a Hybrid genomes and genome stabilization - This section is very lengthy. I would suggest to split it into 2 parts depending on the outcome. For example, the first three paragraphs in this section could be kept, and a second section called “Genome-wide consequences of hybridization” (or something similar) could be created, corresponding to the last 4 paragraphs there. 1.8b - “recombination is suppressed near genes” - I find this somewhat misleading as there is no mechanistic reduction of recombination - what ref 51 shows is that the ancestry blocks are on average longer. Recombination events are not rarer - this reduction in effective recombination rate stems from individuals not recombining having on average a higher fitness. I would recommend to probably rephrase to “effective recombination rate is reduced, due to the lower fitness of the recombinants, more likely to carry incompatible alleles along the same chromosome”. This sentence confused me because at first I understood that there was selection for a reduction in recombination rate (modifier model) around DMIs. 1.8c - “the time to genome stabilization is variable”. In a simple model, Blanckaert & Bank 2018 showed that this time is highly influenced by population size. - “the balance between allelic … what introgression tract will be inherited”. The relative frequency of the parental genotype is a major force here (because of frequency dependent selection) Genetic drift as well as the linkage architecture also plays a role. This is especially true for hybrid species. I would recommend to add these two effects to the paragraph for completeness.


 * Response: We have now added a header “Altered genome properties of hybrid taxa” splitting the section in two parts as suggested by the reviewer. In response to b) we have replaced “recombination” by “the breakdown of ancestry blocks through recombination” which better captures what we intended to say and have added “due to lower fitness of recombinant hybrids” as the reviewer suggests. In response to c) we now cite Blanckaert and Bank (2018) alongside Comeault (2018) and Schumer (2015) as simulation studies showing important factors for hybrid speciation through sorting of incompatibilities.

1.9 Conclusions - “As hybridization is frequent, ...” I do not understand the causality link with the next part of the sentence. Also “hybridization is frequent” has two meanings - either this phenomenon happens for a lot of species pair or lots of hybrid individuals were produced from the hybridization of a given pair. It would be great if the authors could clarify this. - First sentence of the second paragraph: I would recommend to change to “....evidence that … is common, contributes novel species and ...”


 * Response: We have now revised the section to avoid these issues “As hybrid genomes are frequently observed, the advantage of novel adaptive trait combinations often override potential negative effects from incompatibilities and may enable hybrid lineages to purge these incompatibilities during the process of genome stabilization”. and thank Dr. Blanckaert for this advice. The causality is that since many stable hybrid genomes form, there are advantages to it.

1.10 Figure 1: One outcome of hybridisation in Fig 1 is missing, namely successful invasion of one species by the other and I would recommend to add it to the figure. Although invasion can happen without hybridization, loss of one species is still a possible outcome, as hybridization creates a fitness cost on both parental species (DMI - but also search cost if assortative mating).


 * Response: We agree and have added this option to Fig. 1.

1.11 Figure 2: Homoploid speciation and F1 hybrid are written on top of each other Choice of the polyploid illustration is slightly confusing - as having the F1 hybrid between parental species infertile is not the common case.


 * Response: We have now revised the text to clarify that there are other well-known examples of intermediate sterile hybrids in allopolyploid plant speciation such as Spartina anglica, Senecio cambrensis, S. vulgaris var. hibernicus, and S. eboracensis, but that allopolyploids can also form via fertile intermediates (e.g., Tragopogon).

1.12 Figure 3: Unless I am mistaken, the stack of haplotypes here represents a population and not a genotype (as in Figure 2). I would recommend to use a different representation to make it more clear. I was wondering whether the authors intend to keep this file (nomination for deletion from one of the authors).


 * Response: We apologize for the nomination for deletion and have now fixed this. We do think that our representation is adequate, but agree that we need to clarify and have now included this sentence in the figure legend ” Fd is estimated between a hybrid population and the red parent species, and the haplotypes illustrate example individuals in these populations”.

1.13 Glossary - hybrid genome: form -> from - postzygotic incompatibilities: The authors shortly define first postzygotic incompatibilities as intrinsic postmating barriers and then the definition is expanded to include extrinsic barriers as well. I would recommend to replace the first sentence by: “genetic barrier due to the offspring of a mating being unfit or infertile”.


 * Response: We agree and have now revised the glossary accordingly.

Reviewer 2: Gonzalo Nieto Feliner
2.1 Summarizing all the relevant aspects of a complex and wide-ranging topic as hybridization is not easy, particularly in what refers to its creative role in evolution. In my opinion, the article is balanced, covers all relevant topics and is thus useful. I only have a few suggestions. While it is classic to distinguish adaptive introgression and homoploid hybrid speciation as different outcomes of hybridization, based on some of the studies on Heliconius butterflies I suggest that the possibility that introgressed lineages acquire reproductive isolation and give rise to a hybrid species is considered in the ms.


 * Response: We agree, and have clarified that even introgression of smaller sections of the genome may generate new species. Specifically, we have added the sentence “If traits important for species recognition or reproductive isolation introgress into a population of another species, the introgressed population may become reproductively isolatated against other populations of the same species. Examples of this include Heliconius butterflies, where selective introgression of wing pattern genes between diverged lineages occurs, see e.g. [7], and wing patterns contribute to reproductive isolation in some species pairs with low (e.g. between H. t. florencia and H. t. linaresi) and intermediate levels (e.g. H. c. galanthus/H. pachinus) of divergence [8].”

2.2 “Evolutionary outcomes of hybridization”: Reinforcement could be mentioned as one possible outcome of hybrids that are not viable or sterile. For instance, Phlox drummondii (Hopkins, New Phytol. 2013). Under this same heading or under the following one (“adaptive introgression”), perhaps you could mention the possibility of neutral (non-adaptive) introgression.


 * Response: Definitely. We have now added two sentences on reinforcement (If hybrids are not viable or sterile, hybridization may reduce the reproductive success of the parent species [3,9]. This could potentially lead to reinforcement, selection to strengthen premating isolation [2] and rewritten the text to reflect that introgression can indeed be neutral too ” introgressive hybridization may enable neutral or selectively beneficial alleles to be transferred across species boundaries”.

2.3 “for most suggested examples of homoploid hybrid speciation, the genetic basis of reproductive isolation is still unknown”. There are others in which external isolation appears to have done the job (e.g., Senecio squalidus, in which hybrids were brought to the UK by the beginning of the 18th century away from their progenitors)


 * Response: Thanks for bringing this to our attention. We have now included a sentence explaining how external isolation may result in reproductive isolation (”Strong extrinsic pre-zygotic have been shown to isolate the hybrid species Senecio eboracensis from its parent species, where hybrids are virtually absent in the wild, although a fraction of hybrid offspring are fertile in lab experiments [10]. Lowe & Abbott [10] conclude that selfing, timing of flowering and characters involved in pollinator attraction likely contribute to this external isolation.”) and referred to the work on Senecio eboracensis.

2.4 “Hybrid species can occupy an ecological niche different to those of the parents and may be isolated from the parent species primarily through pre-mating barriers (introgressive speciation, c.f. [30])”. I am not aware of the use of such term (introgressive speciation) in that sense. I assume you refer to what Grant called “hybrid speciation with external barriers”. Incidentally reference #30 appears as a mix of two: the classic Verne Grant’s Plant Speciation (1981) and a paper by Payseur and Rieseberg (2016) A genomic perspective on hybridization and speciation.


 * Response: We agree that citing the Verne Grant paper as “hybrid speciation with external barriers” is better and have now done so. This solves any potential mix-up of references.

2.5 “A similar pattern was found in Galapagos finches where a specific hybrid beak shape and song are thought to have been important [8]”. In a section on reproductive isolation I suggest removing the reference to the beak shape since for reproductive isolation it’s song what is relevant.


 * Response: Indeed. Done. We now write “A similar pattern was found in Geospiza Galapagos finches where a specific hybrid song resulted from the transgressive beak morphology”.

2.6 “In what taxa is hybridization common?” Maybe other potential differences between plants and animals that could underlie the different % of hybrids estimated by Mallet in the two groups could be mentioned. E.g. reproductive isolation (Widmer et al., 2009, Heredity)


 * Response: We agree that other potential differences in the patterns of reproductive isolation between plants and animals should be mentioned. To improve the clarity of this section, we have now divided the section into one part focussing on the frequency of hybridization, and another on the role of reproductive isolation against parent species. We have mentioned that the higher frequency of selfing in plants may reduce the rate of hybrid formation by reducing outcrossing opportunities.

2.7 “Paradoxically, the ability to self-pollinate may also act in favour of stabilising allopolyploid taxa”. This is also the case in recombinational speciation.


 * Response: We have now clarified “Selfing is also expected to increase the likelihood of establishment for homoploid hybrids [12]”.

2.8 “Recombinatorial hybrid speciation: the process by which a hybrid taxon develops reproductive isolation from the parent species through sorting of incompatibilities”. Here and elsewhere I suppose you want to say “Recombinational” (cf. Grant, 1971). In addition, I suggest saying “a hybrid population” instead of “a hybrid taxon” since the population(s) developing reproductive isolation need not yet be considered a taxon.


 * Response: Indeed. This is revised in the updated version.

2.9 There are also a number of typos across the text as well as some inconsistent use of UK and US English spelling and some missing words (e.g., in the Box on genomic tools, “Other phylogenetic network methods that [something missing here] incomplete lineage sorting and hybridization may also help [131,132].”


 * Response: Our apologies for this. We have gone over the text and done our best to correct this. We have corrected the sentence by adding “account for” to where the reviewer correctly suspected something to be missing.

Reviewer 3: James Mallet
I thought this was a very full and thorough review. I enjoyed reading it. I'd probably have written it somewhat differently myself in places, but it is very useful just the way it is.

I question the worth of doing this as a Wiki page, however. If I were Runemark et al., I think I would want this to be a proper published review article rather than a Wiki page. I use Wikipedia a lot, and indeed all of us probably do as a first port of call to find out quickly about stuff. But the point of Wikipedia is (a) that it is UNauthored; in other words, anyone can weigh in on it and change stuff they don't like, and (b) it has to be fairly brief for a different more public market than a scientific journal review article. Also, a review article should be a statement of the understanding at the time it was written. As knowledge advances, the review article will go out of date, and it's probably best to write a new review article rather than restructure the old one. So, I would have thought that this half-way Wiki/review article is not going to be so useful... A Wikipedia page is sufficiently flexible that it can change. But of course, I could be wrong!


 * Response: We are very happy that Prof. Mallet liked our review. We decided to write this review in response to an invitation to write a topic page on hybrid genomes for Plos Genetics, with joint publication at Wikipedia to reach non-experts as well as biologists. Our primary goal is hence to produce a PLoS Genetics topic review, but we are all very supportive of open access initiatives. The Wiki page will change over time as new results are found and other people modify it. In addition, there will hopefully be the static review published in PLoS Genetics which reflects the understanding at the time of writing.

I've taken advantage of the Wiki format to make a few minor changes to grammar and typos directly on the document that I think the authors will approve.


 * Response: We thank Prof. Mallet for kindly helping us with the editing!

A few extra points and queries I will put here: 3.1 1) "In addition to recombination rate, the coupling ..." I do not understand what is meant here. Coupling is a strange term; I suggest it can be dropped for a more explanatory term. Can you expand and explain?


 * Response: We agree this was unclear, and found the section repeating already made points when reading through. We have also mentioned the phenomenon under the section “Reproductive isolation against parent species”. We have therefore deleted the section containing this specific formulation and hope this has made the text clearer and more concise.

3.2 2) "Minor parent," "minority parent ancestry," etc. In general, we should always entertain the possibility that the donor of introgression may become the majority parent. Such things appear to have happened in the Anopheles gambiae group, and in several unpublished cases in Heliconius. I would prefer if you could use donor and recipient, although in some of the cases you discuss (Xiphophorus and humans) the donor is indeed the minority parent.


 * Response: We agree that terminology does not cover all cases, and we have now revised the terminology to donor and recipient and explained that the majority parent is not always the recipient of introgression citing the Anopheles example. The section now reads “but in other cases they are highly unequal such as in some Heliconius species [13]. The majority ancestry may even be that from the donor of introgressed material, as was shown for Anopheles gambiae mosquitoes [14]”.

3.3 3) "asexual fusion of cells or hyphae". If it's a hybrid, surely it is sex! I have edited the page to read "ameiotic fusion ..." -- I hope that's ok.


 * Response: Indeed. Thanks for editing!

3.4 4) "Primary" and "secondary hybridization:" I really don't like those terms, though I know they have been used in the literature. It's reminiscent of primary and secondary contact, but surely it just refers to the age of the species before hybridization, rather than the geography involved (even though geography may sometimes of course contribute to a period of lack of hybridization).


 * Response: We agree and have now rewritten the relevant section “Genetic exchange can occur between populations or incipient species diverging in geographical proximity or between divergent taxa that come into secondary contact. Hybridization between more diverged lineages is expected to have a greater potential to contribute beneficial alleles or generate novelty than hybridization between less diverged populations because more divergent alleles are combined in secondary hybridization and are thus more likely to have a large fitness effect, to generate transgressive phenotypes[15].”

Reviewer 4: Roger Butlin
4.1 “Hybridization is the process where genetically closely related species mate and produce offspring with admixed genomes.” This is not the most commonly-used definition of hybridization. The term is more often used to describe mating and offspring production between genetically divergent populations. This can include interspecific hybridization if the species concept in use allows for the production of viable and fertile offspring. To restrict the term to crosses between species excludes many interesting cases, especially examples of hybrid zones.


 * Response: This is a good point. We have now specified that this review focuses on interspecific hybridization and hybrid genomes that result from interbreeding across species boundaries, and that we use hybridization to refer to interspecific hybridization in the review. We now clarify this in the last sentence of the Introduction: “Many of the discussed topics also apply to hybridization between different subspecies or populations of the same species, but here we focus on interspecific hybridization (referred to as hybridization in this review).”

4.2 “If hybrids are not inviable or sterile …” ‘Hybrids’ come in many different flavours! First generation hybrids have one copy of each genome (in a typical diploid and excluding sex chromosomes, which might cause important differences between the sexes). For weakly-diverged parents, they may show heterosis and they are unlikely to show incompatibilities. Later generation hybrids have admixed genomes with many genotypes possible and a wide range of possible fitness effects. Therefore, it is dangerous to refer to ‘hybrids’ as a single group without being more specific. In particular, ‘hybrid fitness’ always needs to be qualified to describe the class of hybrids and to consider variation in fitness among genotypes.


 * Response: We agree that this is an important point. We have specified that we refer to early-generation hybrids and written a sentence to explain that hybrid fitness may depend on the cross-type, generation and parental divergence to clarify that this is the case to the readers.

4.3 “Below we describe the evolutionary outcomes of hybridisation that result in hybrid genomes.” I think the authors have in mind ‘persistent hybrid genomes’, where one or a small number of possible combinations of parental haplotypes comes to dominate a population. This is different from a hybrid zone, for example, where ‘hybrid genomes’ containing many different mixtures of parental alleles are continuously generated but also continuously removed by selection.


 * Response: We agree that we do indeed refer to persistent hybrid genomes, and have revised text to reflect this.

4.4 “Some researchers argue that evidence of a hybridization-derived basis for reproductive isolation should be an additional defining criterion of a hybrid species [26], but see [27].” This debate is about ‘hybrid speciation’ rather than about ‘hybrid species’. Schumer et al. [26] would accept that a species with a genome made up of a stable mix of two parental genomes is a hybrid species, without evidence about the origin of reproductive isolation. However, to describe the process giving rise to this species as ‘hybrid speciation’ they would require evidence that reproductive isolation was generated, at least in part, by hybridisation.


 * Response: We agree, and have now changed “hybrid species” to “hybrid speciation” in this case, and whenever applicable in the section referred to.

4.5 “To establish successfully as a hybrid species, reproductive isolation against both parent species is required” Reproductive isolation is a continuous quantity. Complete reproductive isolation is not required for successful establishment but it must be strong enough for the hybrid and parental populations to retain their distinct characteristics.


 * Response: Indeed. We have now rephrased the sentence to ”To establish successfully as a hybrid species, sufficient reproductive isolation against both parent species for the taxa to remain distinct is required [16-18].”.

4.6 “Recombination is more frequent in homoploid hybrid genomes than in allopolyploid hybrid genomes.” Allopolyploids may have other ways to generate diversity and novelty, for example through differential loss or silencing of gene copies.


 * Response: We agree, and we have also mentioned some of these phenomena in the section on allopolyploid hybrid genomes. Diploidization, chromosomal rearrangements from genomic shock or non-homologous recombination, release of transposable element, diploidization and subgenome dominance are all explained in the section under the new heading “Altered genome properties in hybrid taxa”.

4.7 “Genome stabilization” It is important to note that an isolated hybrid population will eventually stabilise for a pattern of homozygous blocks derived from the two parent species in the absence of selection. This can be taken as a kind of null expectation against which to compare observed patterns. In addition to the processes mentioned, many blocks may show heterosis because of the different sets of deleterious recessive alleles inherited from the parental taxa. This can act to slow down stabilization, especially in areas of low recombination.


 * Response: These are two good points. We have now added “Given time, genetic drift will eventually stochastically fix blocks derived from the two parent species in finite isolated hybrid populations[19]. Selection against incompatibility loci may accelerate the process of fixation of parental alleles as hybrids that possess alleles that are less likely to cause incompatibility will have higher fitness and favourable alleles will spread in the population. Fixation of recessive weakly deleterious alleles in the parent taxa may, however, also result in that hybrids retaining both parental alleles and hence are not homozygous for any weakly deleterious alleles have higher fitness than homozygous hybrids. This associative overdominance [20,21], may slow down the process of fixation of parental alleles through favouring retention of both parental alleles. The effect of assortative overdominance is strongest in low recombination regions, including inversions [22]”.

4.8 “In addition to recombination rate, the coupling to incompatible loci may be important for the likelihood of retention of introgressed material” The meaning of ‘coupling’ in this sentence is unclear.


 * Response: We agree this was unclear, and found the section repeating already made points when reading through. We have also mentioned the phenomenon under the section “Reproductive isolation against parent species”. We have therefore deleted the section containing this specific formulation and hope this has made the text clearer and more concise.

4.9 “Schumer and Brandvain found that hybridization-derived regions on average are less constrained than the rest of the genome [54]. During the process of hybrid genome stabilization the minor parent genome content is overrepresented in high recombination areas, both in swordtail fish and in humans [31].” What is meant by ‘less constrained’ here? Why are minor-parent haplotypes overrepresented in high recombination areas (presumably relative to low recombination areas, rather than to some external expectation)? Is it actually that the minor haplotype in underrepresented in low recombination areas because these areas retain incompatibilities that are expressed more often because the minor haplotype is more often in a ‘foreign’ background?


 * Response: We agree this was unclear, and found the section repeating already made points when reading through. We have also mentioned the phenomenon under the section “Reproductive isolation against parent species”. We have therefore deleted the section containing this specific formulation and hope this has made the text clearer and more concise.

4.10 “In what taxa is hybridization common?” This is quite a different question from, ‘In what taxa is formation of hybrid genomes common?’ On current evidence, hybrid zones are much more widespread than formation of stable hybrid genomes. The two outcomes may not be well correlated across taxa.


 * Response: We meant ‘ What factors influence the likelihood of the formation of persistent hybrid genomes?’. We have now revised this heading and rewritten to clarify that rate of hybridization is only one of the factors deciding the outcome. We have also written an introductory sentence explaining that hybridization is only one of the factors needed for the successful formation of stable hybrid genomes that reads “Whereas hybridization is required for the generation of persistent hybrid genomes, it is not sufficient. For the persistence of hybrid genomes in hybrid species they need to be sufficiently reproductively isolated from their parent species to avoid species fusion. Selection on introgressed variants allows the persistence of hybrid genomes in introgressed lineages. Frequency of hybridization, viability of hybrids, and the ease at which reproductive isolation against the parent species arises or strength of selection to maintain introgressed regions are hence factors influencing the rate of formation of stable hybrid lineages”.

4.11 “Fungal hybridization may also result in asexual hybrid species” This is also certainly true for animal speciation.


 * Response: We have now added the sentence “Hybridization between strongly divergent animal taxa may also generate asexual hybrid species as shown e.g. in the European spined loaches, Cobitis [23] and most if not all asexual vertebrate species are of hybrid origin [24]”.

4.12 “Secondary hybridization is expected to have a greater potential to contribute beneficial alleles or generate novelty than primary hybridization because more divergent alleles are combined in secondary hybridization and are thus more likely to have a large fitness effect or to generate transgressive phenotypes” Secondary hybridisation is also more likely to generate incompatibility (either intrinsic or due to maladaptation). Therefore, it is not clear that there is a simple monotonic relationship between divergence and the probability of generating persistent hybrid genomes.


 * Response: We agree. We have rewritten the primary- and secondary hybridization section in response to comment 3.4, and have included the sentence “Hybridization between more diverged lineages is also more likely to generate incompatible allele combinations reducing initial hybrid fitness [25] but potentially also contributing to hybrid speciation if they are sorted reciprocally as described above. An intermediate genetic distance may thus be most condusive to hybrid speciation [15]. Experimental lab crosses support this hypothesis [26]” in the revised text.

4.13 “Hybrid genomes, including those of many individuals of our own species” This seems to stretch the definition of a ‘hybrid genome’. The term loses its usefulness if it can describe an individual’s genome that happens to contain alleles derived from distinct ancestral populations. Earlier in thus article, it was used to describe stabilized genomes typical of populations. Unless there are Neanderthal or other haplotypes fixed for some parts of the human genome, the term should not be applied.


 * Response: We agree to this point and have now removed the section of the sentence on humans.

4.14 “Box: Detecting and studying hybridization with genomic tools” It would be helpful to distinguish between detection of hybridisation and detection of persistent hybrid genomes.


 * Response: Indeed. We have now revised the box to reflect this. We have expanded the paragraph on dating of admixture events with additional methods to test levels of genome stabilization. We now write: Methods based on linkage disequilibrium decay or methods inferring ancestry tracts can be used to date recent admixture or introgression events as over time ancestry tracts are continuously broken down by recombination [27-32]. With increasing genome stabilization, individuals should vary less in local ancestry. Levels of genome stabilization can thus be assessed by computing the ancestry proportions (e.g. with fd [33]) in genomic windows and testing if these correlate across individuals. Additionally, if hybridization is still ongoing, ancestry proportions should vary across individuals and in space.


 * We thank the reviewers for constructive comments that have helped us improve the manuscript!

(only these referred to in the revised text sections for reference, the numbering is not consistent with that on the wiki-page for simplicity)
 * References

1.	Vallejo-Marín M, Hiscock SJ. Hybridization and hybrid speciation under global change. New Phytol. John Wiley & Sons, Ltd (10.1111); 2016;211: 1170–1187. doi:10.1111/nph.14004

2.	Servedio MR, Noor MAF. The Role of Reinforcement in Speciation: Theory and Data. Annu Rev Ecol Evol Syst. Annual Reviews 4139 El Camino Way, P.O. Box 10139, Palo Alto, CA 94303-0139, USA; 2003;34: 339–364. doi:10.1146/annurev.ecolsys.34.011802.132412

3.	Wolf DE, Takebayashi N, Rieseberg LH. Predicting the Risk of Extinction through Hybridization. Conservation Biology. John Wiley & Sons, Ltd (10.1111); 2001;15: 1039–1053. doi:10.1046/j.1523-1739.2001.0150041039.x

4.	Barton N, Bengtsson BO. The barrier to genetic exchange between hybridising populations. Heredity. Nature Publishing Group; 1986;57 ( Pt 3): 357–376. doi:10.1038/hdy.1986.135

5.	Demon I, Haccou P, van den Bosch F. Introgression of resistance genes between populations: A model study of insecticide resistance in Bemisia tabaci. Theoretical Population Biology. 2007;72: 292–304. doi:10.1016/j.tpb.2007.06.005

6.	Uecker H, Setter D, Hermisson J. Adaptive gene introgression after secondary contact. J Math Biol. Springer Berlin Heidelberg; 2014;70: 1523–1580. doi:10.1007/s00285-014-0802-y

7.	Kronforst MR, Papa R. The Functional Basis of Wing Patterning in Heliconius Butterflies: The Molecules Behind Mimicry. Genetics. Genetics; 2015;200: 1–19. doi:10.1534/genetics.114.172387

8.	Mérot C, Salazar C, Merrill RM, Jiggins CD, Joron M. What shapes the continuum of reproductive isolation? Lessons from Heliconiusbutterflies. Proceedings of the Royal Society B: Biological Sciences. The Royal Society; 2017;284: 20170335. doi:10.1098/rspb.2017.0335

9.	Prentis PJ, White EM, Radford IJ, Lowe AJ, Clarke AR. Can hybridization cause local extinction: a case for demographic swamping of the Australian native Senecio pinnatifolius by the invasive Senecio madagascariensis? New Phytol. John Wiley & Sons, Ltd (10.1111); 2007;176: 902–912. doi:10.1111/j.1469-8137.2007.02217.x

10.	Lowe AJ, Abbott RJ. Reproductive isolation of a new hybrid species, Senecio eboracensis Abbott & Lowe (Asteraceae). Heredity. Nature Publishing Group; 2004;92: 386–395. doi:10.1038/sj.hdy.6800432

11.	Widmer A, Lexer C, Cozzolino S. Evolution of reproductive isolation in plants. Heredity. 2009;102: 31–38. doi:10.1038/hdy.2008.69

12.	McCarthy EM, Asmussen MA, Anderson WW. A theoretical assessment of recombinational speciation. Heredity. Nature Publishing Group; 1995;74: 502–509. doi:10.1038/hdy.1995.71

13.	Jiggins CD, Salazar C, Linares M, Mavárez J. Review. Hybrid trait speciation and Heliconius butterflies. Philos Trans R Soc Lond, B, Biol Sci. 2008;363: 3047–3054. doi:10.1098/rstb.2008.0065

14.	Fontaine MC, Pease JB, Steele A, Waterhouse RM, Neafsey DE, Sharakhov IV, et al. Extensive introgression in a malaria vector species complex revealed by phylogenomics. Science. American Association for the Advancement of Science; 2015;347: 1258524–1258524. doi:10.1126/science.1258524

15.	Marques DA, Meier JI, Seehausen O. A combinatorial view on speciation and adaptive radiation. Trends in Ecology & Evolution. University of California Press; 2019.

16.	Abbott RJ, Rieseberg LH. Hybrid Speciation. Chichester, UK: John Wiley & Sons, Ltd; 2001. doi:10.1002/9780470015902.a0001753.pub2

17.	Abbott R, Albach D, Ansell S, Arntzen JW, Baird SJE, Bierne N, et al. Hybridization and speciation. J Evol Biol. 2013;26: 229–246. doi:10.1111/j.1420-9101.2012.02599.x

18.	Schumer M, Rosenthal GG, Andolfatto P. HOW COMMON IS HOMOPLOID HYBRID SPECIATION? Evolution. 2014;68: 1553–1560. doi:10.1111/evo.12399

19.	Buerkle CA, Rieseberg LH. The rate of genome stabilization in homoploid hybrid species. Evolution. 2008;62: 266–275. doi:10.1111/j.1558-5646.2007.00267.x

20.	Ota T. Associative overdominance caused by linked detrimental mutations. Genetical Research. 1971;18: 277–286.

21.	Zhao L, Charlesworth B. Resolving the Conflict Between Associative Overdominance and Background Selection. Genetics. Genetics; 2016;203: 1315–1334. doi:10.1534/genetics.116.188912

22.	Faria R, Johannesson K, Butlin RK, Westram AM. Evolving Inversions. Trends in Ecology & Evolution. 2019;34: 239–248. doi:10.1016/j.tree.2018.12.005

23.	Janko K, Pačes J, Wilkinson-Herbots H, Costa RJ, Roslein J, Drozd P, et al. Hybrid asexuality as a primary postzygotic barrier between nascent species: On the interconnection between asexuality, hybridization and speciation. Mol Ecol. 2017;27: 248–263. doi:10.1111/mec.14377

24.	Neaves WB, Baumann P. Unisexual reproduction among vertebrates. Trends in Genetics. 2011;27: 81–88. doi:10.1016/j.tig.2010.12.002

25.	Maheshwari S, Barbash DA. The Genetics of Hybrid Incompatibilities. Annu Rev Genet. 2011;45: 331–355. doi:10.1146/annurev-genet-110410-132514

26.	Comeault AA, Matute DR. Genetic divergence and the number of hybridizing species affect the path to homoploid hybrid speciation. Proceedings of the National Academy of Sciences. 2018;115: 9761–9766. doi:10.1073/pnas.1809685115

27.	Moorjani P, Patterson N, Hirschhorn JN, Keinan A, Hao L, Atzmon G, et al. The History of African Gene Flow into Southern Europeans, Levantines, and Jews. PLoS Genet. 2011;7: e1001373. doi:10.1371/journal.pgen.1001373

28.	Moorjani P, Sankararaman S, Fu Q, Przeworski M, Patterson N, Reich D. A genetic method for dating ancient genomes provides a direct estimate of human generation interval in the last 45,000 years. Proc Natl Acad Sci USA. 2016;113: 5652–5657. doi:10.1073/pnas.1514696113

29.	Patterson N, Moorjani P, Luo Y, Mallick S, Rohland N, Zhan Y, et al. Ancient admixture in human history. Genetics. 2012;192: 1065–1093. doi:10.1534/genetics.112.145037

30.	Loh P-R, Lipson M, Patterson N, Moorjani P, Pickrell JK, Reich D, et al. Inferring Admixture Histories of Human Populations Using Linkage Disequilibrium. Genetics. 2013;193: 1233–1254. doi:10.1534/genetics.112.147330

31.	Sankararaman S, Patterson N, Li H, Pääbo S, Reich D. The date of interbreeding between Neandertals and modern humans. PLoS Genet. 2012;8: e1002947. doi:10.1371/journal.pgen.1002947

32.	Meier JI, Marques DA, Mwaiko S, Wagner CE, Excoffier L, Seehausen O. Ancient hybridization fuels rapid cichlid fish adaptive radiations. Nature Communications. Nature Publishing Group; 2017;8: 1–11. doi:10.1038/ncomms14363

33.	Martin SH, Davey JW, Jiggins CD. Evaluating the Use of ABBA-BABA Statistics to Locate Introgressed Loci. Molecular Biology and Evolution. 2014;32: 244–257. doi:10.1093/molbev/msu269

Second Review by Reviewer 1: Alexandre Blanckaert
ID version: 7010

I want to thank the authors for answering my comments. I enjoyed reading the revised manuscript. The authors addressed the comments from all four reviewers. I believe that the article can be published in its current form.

I have three minor comments remaining: general: the author used indifferently pre-zygotic and prezygotic. Abstract: I would add genomes in the following sentence “ … which are a mosaic of the parent species genomes with no increase in chromosome number” For Fig 2, I still have some text overlapping each other: namely “Homoploid hybrid and F1 hybrid”. In addition, there is two species names below the picture of the second sparrow on the left. One is Passer hispaniolensis, the other is just above and not readable.

I have noticed a few formatting problems/typos: abstract:
 * something happened to the word homoploid

Evolutionary outcome of hybridization:
 * extra space after ref [27,28].

Adaptive introgression:
 * the parenthesis opened in the second sentence is not closed.

What is a hybrid species:
 * last sentence: variation exists

Homoploid and polyploid hybrid speciation:
 * after the second (Fig. 2). there is an extra ).

Reproductive isolation against parent species:
 * extra space after ref [52].
 * second last sentence I would add “/or” “if mating is time and/or habitat specific”.
 * Add space between species and [65]
 * Add space between species and [66]

Altered genome properties in hybrid taxa:
 * merge ref [102],[105] into [102,105]

What factors influence the likelihood ….:
 * The quote used to explain Haldane’s rule is lacking a starting point.
 * merge ref [121][102] into [102,121]
 * add space between subspecies and [34]

Second Review by Reviewer 4: Roger Butlin
I agree with Alexandre that the authors have done a great job in revising this article in response to reviewer comments. I will make a very useful contribution to the field.

I have no further substantive comments, just a few editorial suggestions.

In the section "Stabilization of hybrid genomes", I suggest replacing, “Fixation of recessive weakly deleterious alleles in the parent taxa may, however, also result in that hybrids retaining both parental alleles and hence are not homozygous for any weakly deleterious alleles have higher fitness than homozygous hybrids. This associative overdominance [76,77], may slow down the process of fixation of parental alleles through favouring retention of both parental alleles. The effect of assortative overdominance is strongest in low recombination regions, including inversions [78].” with: “Fixation of recessive weakly deleterious alleles in the parent taxa may, however, also result in hybrids retaining both parental alleles: because hybrid s with haplotypes from both parents are not homozygous for any weakly deleterious alleles, they have higher fitness than hybrids with only one parental haplotype. This associative overdominance [76,77], may slow down the process of fixation of parental alleles through favouring retention of both parental haplotypes. The effect of associative overdominance is strongest in low recombination regions, including inversions [78].”

It would be good to italicize formal names throughout.

Please replace ‘artic floras’ with ‘Arctic floras’. [Actually, a general proof-read would be helpful - Alexandre has picked up some issues but there are some others.]

Response letter second review
We have now revised the manuscript according to the comments of both referees as well as exchanged the figure file format for .svg and added wikipedia links throughout the manuscript. We hope that this manuscript would now be suitable for publication as a topic page in Plos Genetics.

Author response - Second Review by Reviewer 1: Alexandre Blanckaert
ID version: 7010

I want to thank the authors for answering my comments. I enjoyed reading the revised manuscript. The authors addressed the comments from all four reviewers. I believe that the article can be published in its current form.’’

>>Response: We are very happy that Dr. Blanckaert enjoyed reading our manuscript and find it ready for publishing. We will revise the manuscript to meet the remaining concerns.

1:1 I have three minor comments remaining: general: the author used indifferently pre-zygotic and prezygotic. Abstract: I would add genomes in the following sentence “ … which are a mosaic of the parent species genomes with no increase in chromosome number” Response: We agree and have now done so.

1:2 For Fig 2, I still have some text overlapping each other: namely “Homoploid hybrid and F1 hybrid”. >>Response: We have now fixed this issue and provided the figure in svg-format.

1:3 In addition, there is two species names below the picture of the second sparrow on the left. One is Passer hispaniolensis, the other is just above and not readable. >>Response: Thanks for spotting this! We have now revised the figure.

1:4 I have noticed a few formatting problems/typos: abstract: something happened to the word homoploid

Evolutionary outcome of hybridization:

extra space after ref [27,28]. Adaptive introgression:

the parenthesis opened in the second sentence is not closed. What is a hybrid species:

last sentence: variation exists Homoploid and polyploid hybrid speciation:

after the second (Fig. 2). there is an extra ). Reproductive isolation against parent species:

extra space after ref [52]. second last sentence I would add “/or” “if mating is time and/or habitat specific”. Add space between species and [65] Add space between species and [66] Altered genome properties in hybrid taxa:

merge ref [102],[105] into [102,105] What factors influence the likelihood ….:

The quote used to explain Haldane’s rule is lacking a starting point. merge ref [121][102] into [102,121] add space between subspecies and [34]

>>Response: Thanks for a thorough read through and for spotting these mistakes. We have now fixed these issues.

Author response - Second Review by Reviewer 4: Roger Butlin
I agree with Alexandre that the authors have done a great job in revising this article in response to reviewer comments. I will make a very useful contribution to the field.

>> Response: We are very happy that Prof. Butlin thinks our review will make a useful contribution to the field.

I have no further substantive comments, just a few editorial suggestions.

In the section "Stabilization of hybrid genomes", I suggest replacing, “Fixation of recessive weakly deleterious alleles in the parent taxa may, however, also result in that hybrids retaining both parental alleles and hence are not homozygous for any weakly deleterious alleles have higher fitness than homozygous hybrids. This associative overdominance [76,77], may slow down the process of fixation of parental alleles through favouring retention of both parental alleles. The effect of assortative overdominance is strongest in low recombination regions, including inversions [78].” with: “Fixation of recessive weakly deleterious alleles in the parent taxa may, however, also result in hybrids retaining both parental alleles: because hybrid s with haplotypes from both parents are not homozygous for any weakly deleterious alleles, they have higher fitness than hybrids with only one parental haplotype. This associative overdominance [76,77], may slow down the process of fixation of parental alleles through favouring retention of both parental haplotypes. The effect of associative overdominance is strongest in low recombination regions, including inversions [78].”

It would be good to italicize formal names throughout.

Please replace ‘artic floras’ with ‘Arctic floras’. [Actually, a general proof-read would be helpful - Alexandre has picked up some issues but there are some others.]’’

>> Response: We have now revised the manuscript and included all Prof. Butlins editorial suggestions and done our best to proofread the manuscript.