A postmeiotic route to stepwise polyploidy

Article activity feed

1. 

   Review Commons

   Sep 24, 2026

   Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Reply to the reviewers

   Dear Editor and Reviewers,

   We thank the Editor and the three reviewers for their careful, constructive, and encouraging assessments of our manuscript. Their comments helped us improve the accessibility of the text, clarify the experimental logic, expand the methodological details, and temper several claims.

   We believe that the revised manuscript has improved a lot and is now substantially more accessible to a broad audience. In particular, we have removed or minimized specialized terminology and technical labels where they were not essential, replacing them with more straightforward descriptions. We have also reorganized the experimental narrative to …

   Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Reply to the reviewers

   Dear Editor and Reviewers,

   We thank the Editor and the three reviewers for their careful, constructive, and encouraging assessments of our manuscript. Their comments helped us improve the accessibility of the text, clarify the experimental logic, expand the methodological details, and temper several claims.

   We believe that the revised manuscript has improved a lot and is now substantially more accessible to a broad audience. In particular, we have removed or minimized specialized terminology and technical labels where they were not essential, replacing them with more straightforward descriptions. We have also reorganized the experimental narrative to make the underlying logic easier to follow, revised the figure legends, applied consistent color coding, and added explicit explanations of the assays and their interpretation. We have also tempered the central claim, replacing the description of SEM as “the predominant” mechanism with the more appropriate formulation that it represents “a major” route, while explicitly acknowledging alternative explanations. Together, these changes reduce jargon, clarify the reasoning, and make the significance of our findings more accessible to readers beyond the field of yeast genetics.

   The revised manuscript was read by several researchers who are not yeast geneticists, whose feedback confirmed that the text is clear, coherent, and accessible to readers outside the immediate specialist community.

   Reviewer #1

   Evidence, reproducibility and clarity

   Summary This paper analyzes the process of ploidy alterations amongst a set of diverse yeast (S. cerevisiae) strains and documents the sequential appearance of ~diploid, ~ triploid and ~tetraploid isolates from asci. It argues that this is the predominant mechanism of ploidy increases and that it explains many findings in other studies.

   *Major comments

   1. • The key conclusion of the paper is that the series of consecutive events described is "the predominant natural route to polyploidy in yeast". However, the events shown remain relatively rare (and if they are not rare, no comparison to frequency of other events is provided). In addition, the experiments start with a selection of strains picked for their ability to mate. a. A figure illustrating the frequency of each step in the process would help put the results into larger context.* Response: The experiments were performed on different strain sets and were designed to test the individual steps of SEM rather than to estimate a population-wide rate for the complete sequence. A figure illustrating the frequency of each step is not feasible because we did not establish any population level estimate of their rates. The revised manuscript reports the relevant conditional observations in the Results. In the first cycle, all 32 strains in the spore-derived experiment sporulated, five of nine heterothallic strains produced colonies with ploidy consistent with postmeiotic genome doubling, and only two backgrounds (AKR and ALH) generated triploid progeny after mating tests. In the intact-ascus experiment, five of twelve tested backgrounds generated triploid colonies, with observed frequencies of 1.2–9.1% among the relevant colonies. For the second cycle, 38 of 46 natural triploids sporulated; among 32 sequenced triploid-derived colonies, six (19%) displayed the combined genomic signatures of postmeiotic genome doubling, and five of these six remained mating competent. The complete 3x-to-4x transition was directly recovered in the ALH background. We now present these values as conditional experimental frequencies, not as estimates of the natural incidence of SEM. We have therefore tempered the central claim of “predominant route”, explicitly described the strain-selection limitations, and clarified that SEM is one important route among several.

   • While the data show that the series of sporulation/endoreplication/and mating appears to generate cells with altered ploidy and extensive aneuploidy, the evidence that this is "the predominant natural route" is difficult to extract from the work as presented. There are references to prior papers, but the key points in those papers should be presented in a manner that is convincing-perhaps in a table that compares issues addressed in this work and that build the argument made in favor of this SEM mechanism being 'predominant (e.g, frequency of aneuploidy, frequency of triploidy, tetraploidy, ploidy stability etc.)*

   Response: We thank the reviewer for this constructive suggestion. Figure 5 now includes a summary panel (Fig. 5D) comparing the observed natural-polyploid genomic features with the expectations of the stepwise and saltational routes. This comparison clearly shows that several genomic signatures argue against a dominant saltational origin and support a stepwise polyploidization through iterative SEM cycles as a major route to natural polyploidy in S. cerevisiae.

   *a. An example of this is the statement in lines 62-64 that "8-20% of natural isolates are autopolyploid and triploids are at least as frequent as tetraploids" , but Peter et al found Response: We initially sought to provide a general estimate of polyploidy prevalence by integrating data from several large-scale studies. Following the reviewer’s suggestion, however, we now report the specific numbers from the 1,011-genome study, thereby simplifying and strengthening the presentation “In the 1,011-genome study, 91 of the 742 isolates with assigned natural ploidy were polyploid (11.5%), including 46 triploids and 39 tetraploids, indicating that triploids were slightly more common than tetraploids”.

   • Figure 2B is extremely confusing and would benefit by making it easier to understand. a. First, it would help to understand how these particular strains were chosen from the 728 starting strains. The text says that it chose those that could mate, but if step 1 is sporulation, why not test the hundreds of strains that could sporulate (based on Table S2). The left part of Fig 2B illustrates a mating score inferred based on 3 assays. Are the a or alpha designations and listing of hetero vs homo zygosity from before or after the experiment? How part C relates to part B of the figure is unclear, one would need to trace the strains from the left of 2B to the bottom of 2C.*

   Response: We agree and substantially redesigned Figure 2 by reversing the order of the original panels B and C and adding explanatory schematics and clarification within each panel and its legend. The revised figure now uses a consistent color code for haploid, diploid, and triploid ploidy classes, as well as for mating-type composition. Mating scores are explicitly described as composite scores derived from mass-mating assays, direct observation of zygote formation, and the creeping phenotype, which is now clearly defined as a mating-associated aggregation phenotype used as an additional readout of mating competence. The revised text also clarifies that the spore-derived endoreplication experiment, presented in the new Panel B, addresses the first endoreplication step of the SEM cycle (E1), whereas the new Panel C presents the natural-diploid mating screen addressing the first mating step (M1).

   • The title of the legend states that "endoreduplication during spore germination enables diploid to triploid transition" (line 180). Yet, the 'triploids generated are primarily Response: We addressed these questions by redesigning the new Panel B to show more clearly that postmeiotic genome doubling was detected only in non-self-mating (heterothallic) strains, which generated approximately 2x spore-derived colonies. By contrast, as expected following spore microdissection, all 23 self-mating (homothallic) strains produced MATa/MATα diploid colonies, and none showed evidence of genome doubling, which would have resulted in tetraploid colonies. The revised figure and legend also illustrate that the gray-shaded range represents the approximate ploidy expected from genome doubling and that none of the 23 homothallic strains produced colonies with doubled ploidy.

   c. In Fig 2C, there are a few red or blue spots in the homothallic strains, so the term 'exclusively' in line 190 cannot be true. It is also not clear why endoreduplication would expect to generate strains that are the same ploidy level as the parents, unless the intention is that endoreduplication AFTER meiosis is expected to give diploid progeny. Please be more explicit as there are many unstated assumptions here that are difficult to follow. Response: We apologize for the confusion and thank the reviewer for identifying this lack of clarity. As described above, the revised Figure 2 presents the experimental design and results more explicitly, which should resolve most of the potential misunderstandings. We have also removed the term “exclusively” to account for the small number of single-mating-type colonies, most likely generated by loss of heterozygosity at the MAT locus, recovered from homothallic spore-derived progeny and replaced it with the more accurate term “predominantly.”

   • The triploidy to tetraploidy step (starting at line 196) appears to be built upon analysis of 3 euploid triploid isolates (from the 1011 collection, not from experiments in Figure 2). These exhibit very low spore viability (presumably due to chromosome segregation defects known to be prevalent in spores from triploid parents). Accordingly, these meioses gave rise to aneuploid spores. It is argued that any strains with tetrasomic chromosomes must have arisen via endoreduplication. However, highly aneuploid isolates exhibit high levels of chromosome instability, such that missegregation of extra chromosomal pairs would need to be ruled out to reach this conclusion. Furthermore, if the '1.5x' spores that mate carried 1x of most chromosomes and 2x of other chromosomes (e.g., chromosomes 4,5,6) -what rules out the possibility that they mated with another spore that also had 1x of the same chromosomes and 2x of the same ones (e.g., chromosomes 4,5,6) . a. The illustration of this process in Figure 3 is confusing. For example, the text says that the 3X spore in the ascus after E2 is a/alpha/alpha, but the figure says it's only alpha/alpha. Does it have only alpha/alpha because it has only two copies of chr3 despite being ~3X? Assuming it has only MATalpha, it could then mate with a 1.5x spore that had either 1 or 2 copies of Chr3, and therefore was either a/a or alpha/alpha. b. In Fig. 3C, why is the ABR data for ChrV, VI, X, XI, XV and XVI so sparse? c. There is a focus on monosporic asci (i.e., only one spore survives the meiosis). Please explain the rationale for this and the assumptions that underlie it.*

   Response: We agree that a single four-copy chromosome would not establish postmeiotic genome doubling, especially in aneuploid triploid progeny. Our inference is based on a combination of signatures observed together in six isolates: 1. Several chromosomes were present in four copies, although triploid meiosis alone can generate at most three copies of a given chromosome in a spore. 2. Single-copy chromosomes were completely absent, whereas they were common among non-endoreplicated triploid-derived isolates. 3. Four-copy chromosomes showed ABRs near 0.5, consistent with duplication of pre-existing heterozygous two-copy chromosomes. 4. Two-copy chromosomes lacked heterozygous variants, consistent with duplication of chromosomes that were initially present in one copy. This is the reason why ABR data for ChrV, VI, X, XI, XV and XVI are so sparse. 5. The overall genome content was higher in these isolates (2.2–3.0x; mean 2.6x) than in the other triploid-derived isolates (1.1–2.4x; mean 1.7x), and they remained capable of mating, excluding that they originated from self-mating. The coordinated occurrence of these features is not readily explained by independent chromosome missegregation events. We nevertheless use “putatively endoreplicated” and “consistent with genome doubling” where appropriate, because the molecular mechanism cannot be directly observed retrospectively. We revised Figure 3 to clarify the chromosome-copy-number and ABR logic. The parental genotype shown in the schematic is now stated explicitly, and the revised legend explains that ABR is the fraction of sequencing reads carrying the alternative allele at heterozygous sites. We also clarified that the experiment analyzes three natural triploid backgrounds (BBT, BAD, and CRE).

   • Fig 4 looks at the whole ascus, rather than selected spores. And asks about their DNA content (4A), presumably after sporulation and then whatever mating or endoreduplication+ mating that goes on in the privacy of the ascus.... Most spores were haploid (didn't mate) or diploid (mated with ascus siblings) and 1-9% of the spores were triploid (with 2/8 analyzed having some tetrasomic and/or disomic chromosomes). While this is consistent with the model being proposed, can we rule out the possibility of a partial meiosis that yielded 2 1x spores and 1 2x spore and then mating between the 2x and 1x spores... There may be good arguments for this, but they were difficult to find.*

   Response: There are two main reasons why we consider the occurrence of triploid colonies among the meiotic progeny of diploid parental strains to be more consistent with completion of a first SEM cycle than with partial meiosis. First, meiotic restitution, defined as the omission or incomplete execution of one of the two meiotic divisions, would generate two unreduced diploid gametes within the ascus; mating between two such gametes would produce tetraploid rather than triploid colonies. Second, in the experiment shown in Figure 3, we identified genome-doubled colonies derived from asci containing three or four viable spores, making it unlikely that the observed genome doubling resulted from meiotic chromosome non-reduction. We chose not to include these additional explanations in the main text in order to preserve the manuscript’s narrative flow.

   6. It apparently assumes that 3X spores arose only by endoreduplication and mating between sister spores. a. One confusing thing is that here it shows 2x and 3x progeny, but if we already know that triploid sporulation gives mostly ~1.5x spores (Figure 3) why are they labeled as 1x or 2x inside the asci in Figure 4B?

   Response: Figure 4B depicts progeny derived from a diploid parental cell and therefore does not contain a 2x spore. The reviewer may instead be referring to Figure 4C. We agree that triploid meiosis produces spores with an average ploidy of approximately 1.5x; however, triploid-derived spores display a broad range of chromosome complements because of the uneven segregation characteristic of triploid meiosis, as illustrated in Figure 3B. We therefore replaced the 1x and 2x labels in Figure 4C with ∼1x and ∼2*x *to reflect the variable chromosome complements of triploid-derived spores. We have also extended this notation to near-triploids and near tetraploids and revised the text accordingly to explain that near-haploid and near-diploid colonies likely result from uneven chromosome segregation during triploid meiosis without subsequent mating.

   7. The text is written for a highly specialized audience fluent in the fine points of S. cerevisiae mating, meiosis and ploidy change. Some simple explanations (e.g., homo vs heterothallism and how that is determined -presumably based on Ho gene function but not stated as such; 'bisexuality' - how does this differ from same sex mating and from mating of the same organism with its progeny that has switched to the opposite mating type?) What does dioecy mean for a yeast cell? These questions may seem naïve, but require extensive familiarity with the field; explaining the intended meaning would benefit allow a broader audience of readers to appreciate the fine points of this work.

   Response: We agree and have revised the manuscript to make it accessible to a broader audience. We now introduce the canonical yeast life cycle in the Introduction, describe self-mating and non-self-mating strains in functional terms, and clarify that MATa and MATα designate the two mating types. We have removed specialized terms such as “bisexuality” and “dioecy” and made extensive efforts to simplify the narrative, clarify the experimental logic, and explain technical concepts so that the manuscript can be readily understood by readers who are not specialists in yeast genetics.

   8. Arguments in the paragraph from line 388-398 are hard to follow. Some visual to bolster the explanation and help readers understand why stepwise is definitely more prevalent than saltation in all strains and conditions is needed.

   Response: We agree. We reorganized the Results section around explicit predictions of the two models, including triploid frequency, association with self-fertility, genome-wide homozygosity, ABR profiles, and aneuploidy. We also added a summary panel to Figure 5 comparing the genomic features observed in natural polyploids with the expectations of the stepwise and saltational routes. This comparison makes clear that the genomic properties of natural polyploids are more consistent with a stepwise origin than with a predominantly saltational one, while acknowledging that alternative mechanisms may also contribute.

   9. Lines 404-407-is this true for the natural tetraploids in the 1011 collection as well (that they lack 0.5 ABR peaks?) . Which stains are artificial vs natural vs from industrial fermentation sources.

   Response: The 1,011-collection strains analyzed in this section are all classified as natural isolates as defined by the original study, regardless of their domesticated or industrial status. They are natural by opposition to lab-generated polyploids. Natural tetraploids do not lack an approximately 0.5 ABR peak. Rather, all 38 natural tetraploids display approximately 0.25, 0.5, and 0.75 peaks. The key observation is the presence of the additional 0.25 and 0.75 classes together with the 0.5 class. These patterns are also seen in the experimentally generated ALH-4x isolate and are expected when four-copy chromosomes contain 1:3, 2:2, and 3:1 allele ratios. We added the following sentence “By contrast, natural tetraploids displayed additional ABR peaks at ~0.25, and ~0.75, together with the ~0.5 peak (Fig. 5B), a pattern also observed in the ALH-4x isolate generated through two successive SEM cycles (Fig. 4E). The additional peaks at ~0.25 and ~0.75 are inconsistent with simple WGD of a heterozygous diploid and instead supports a stepwise origin through the SEM sequence.”.

   10. Paragraph 408-416-The argument is not easy to follow-if tetraploids are not so stable, then there should be lots of aneuploidy derived from them as well. Are there publications that have followed chromosome loss from triploid or tetraploids strains?

   Response: This paragraph also was extensively rewritten to simplify the message. Moreover, we now refer to the publication by Mayer & Aguilera (1990) reporting a higher rate of chromosome loss in tetraploids than in triploids.

   • While this paper may have established SEM as a mechanism of generating polyploidy and aneuploidy, it is difficult to understand if this mechanism is as 'predominant' as claimed. Please provide a figure or table to bolster this point. Furthermore, even if SEM is a frequent event, that doesnot rule out other mechanisms, the WGD being one of them. Thus, the authors should temper their statements to allow for other mechanisms that they show do also generate polyploids and aneuploids.*

   Response: We fully agree. The manuscript now consistently uses “a major route” rather than “the predominant route”. We explicitly state that WGD, meiotic restitution, MAT-locus homogenization, and other mechanisms may contribute to natural polyploidization. Our conclusion is that SEM explains a substantial and previously underappreciated fraction of the observations, not that SEM is the only route. A novel panel D was added to Figure 5 to summarize the genomic signatures observed in natural polyploids and their agreement or disagreement with predictions of the stepwise and saltational routes to polyploidization.

   12. Also, was the stepwise process recapitulated starting from haploids to the tetraploids for the same strains? My impression is that at the triploid stage more and 'better' triploids were chosen from the 1011 collection, so has the continuity of the process really been demonstrated definitively?

   Response: We cannot initiate the SEM sequence from a haploid strain because haploid cells do not undergo sporulation. The complete experimental continuity demonstrated here therefore begins with diploid parental strains: diploid-to-triploid formation was observed using intact asci from multiple diploid genetic backgrounds, and experimentally generated triploids were subsequently used to demonstrate the triploid-to-tetraploid transition, including the ALH-3x-to-ALH-4x lineage.

   The three natural triploids selected from the 1,011-genome collection were used specifically to characterize the second SEM cycle and to identify postmeiotic genome-doubling events among their meiotic progeny. They were not selected because they were considered “better” or more representative triploids, but because they were euploid and sporulation competent. The data nevertheless provide direct experimental continuity from diploidy to triploidy and from triploidy to tetraploidy.

   • Finally, how much does strain background play a role in the results and would attempting to repeat these experiments with different strains be likely to yield different results that would be simply ascribed to 'strain-specific effects'? The effort to look at many strains is admirable; explanations of why specific strains were chosen for the work over others would be helpful.*

   Response: We thank the reviewer for recognizing the effort made to analyze multiple strain backgrounds. We agree that genetic background may influence all stages of the SEM sequence, including sporulation, spore viability, postmeiotic genome-doubling propensity, mating competence, and tolerance of aneuploidy. We therefore do not consider the frequencies measured here to be universal, and repeating these experiments with additional strains could yield different outcomes that reflect genuine biological variation rather than merely experimental noise.

   The spore-derived experiment included 32 diploid strains with defined sexual behaviors, comprising 23 self-mating and 9 non-self-mating backgrounds. The intact-ascus experiment deliberately focused on 12 non-self-mating diploid strains because the preceding experiment detected postmeiotic genome doubling in this class and because loss of self-fertility is enriched among natural polyploids. These strains nevertheless represented diverse genetic and ecological backgrounds, and triploid progeny were recovered from several independent backgrounds, indicating that the process is not restricted to a single strain.

   We now explicitly acknowledge in the Discussion that natural variation in genes involved in spindle pole body duplication, chromosome segregation, cytokinesis, cell-cycle control, and aneuploidy tolerance may contribute to differences among strain backgrounds in the frequency and efficiency of the complete SEM sequence. Thus, strain background is an important determinant of the process and a subject for future investigation, while the recovery of SEM-associated outcomes across multiple backgrounds supports the general relevance of the mechanism.

   Minor comments

   Are the strains selected for these studies truly 'wild' isolates or from domesticated yeasts? Could there be a difference between how domesticated/industrial yeasts and wild yeasts become polyploid? How do the strains study here compare with those from previous studies of polyploidization?* Response__:__ We agree that the term “natural polyploids” required clarification. In the revised manuscript, we explicitly state that this category includes both wild isolates recovered from natural environments and domesticated isolates originating from industrial or other human-associated environments. These non-laboratory isolates are distinct from laboratory-derived polyploids generated through experimental genetic manipulation. We now make this distinction explicit in the Introduction and use “natural polyploids” in this broad sense throughout the manuscript.

   With respect to the mechanism of polyploidization, our data do not reveal clear differences between wild and domesticated isolates. Both groups display genomic features compatible with the SEM model, including loss of self-fertility, aneuploidy, and complex allele-balance-ratio profiles. Notably, the ALH strain, in which we experimentally demonstrated the complete SEM sequence, was isolated from horse dung in Ecuador. Other polyploid strains used in our study, such as the triploid strains BBT and BAD, originated from domesticated environments, whereas CRE was isolated from a termite mound. Overall, our results do not suggest that ecological origin is a major discriminant of the mechanism underlying polyploid formation. However, because the evidence supporting this conclusion is limited, we decided not to discuss this aspect in the main text and have restricted it to the response to the reviewer.

   • One striking result that needs to be explained is why growth of a haploid lab strain (Ho deleted, one mating type) led to extremely rapid autodiploidization (mating type homozygous) that provides a growth advantage on glucose limiting medium (https://doi.org/10.1016/j.cell.2016.08.002). How frequently would that type of event happen when cells are stressed a bit?*

   Response: We agree that this is an interesting point. In the Results section, we note that previous studies showed that endoreplication occurs frequently during haploid vegetative growth, with approximately 17% of cells spontaneously becoming diploid after 100 generations in the absence of external stress, and that this frequency can reach 100% in the presence of ethanol or potassium chloride (31). We have now added the reference suggested by the reviewer to note that rapid autodiploidization was also observed during experimental evolution under glucose limitation, where it conferred a growth advantage. This finding further supports the idea that diploidization can arise during adaptation to nutrient-limited environments and may be strongly favored under such conditions (lines 163–165).

   It would be interesting to examine whether stress affects the rate of postmeiotic endoreplication. However, systematically addressing this question is beyond the scope of the present manuscript and will likely require a dedicated follow-up study.

   • Lines 353-356-were sufficient numbers of tetraploids tested to make this a fair comparison?*

   Response: We sequenced 32 colonies derived from individual spores produced by triploid meiosis and detected postmeiotic endoreplication in 6 cases (18.8%; Fig. 3B). In contrast, none of the 28 colonies derived from tetraploid meiosis showed evidence of endoreplication (0%). A two-sided Fisher’s exact test indicated that this difference was statistically significant (p=0.026p = 0.026p=0.026), suggesting that postmeiotic endoreplication occurs more frequently among triploid-derived than tetraploid-derived spores. To preserve the flow of the main text, we have not included this statistical test in the manuscript.

   • In figure 5C, what is the difference between an a mater, a mater and a bisexual mater and, related to this, what is dioecy in the context of yeast mating? It was previously described (ref 41) in polyploid fermentation strains. Can some context be provided on its role here?*

   Response: We understand that the reviewer was referring to Supplementary Fig. 5C rather than Fig. 5C. We have therefore removed this panel and the sentence referring to it to avoid further complicating the narrative with issues related to bisexuality and dioecy. We thank the reviewer for pointing out that these aspects were not essential to the central message of the paper and that their inclusion made the story unnecessarily complex and less accessible to a broad audience.

   • Lines 370-372-the order of the models in Fig1 A are saltation and then stepwise but are discussed here in the reverse order.*

   Response: We thank the reviewer for pointing out this inversion. We have corrected the main text so that the two models are now cited in the order in which they are presented in Figure 1.

   • Chromsome loss from tetraploids referred to from ref 30: Where in this paper is the stability of ploidy levels noted? In scanning all the figures, this reviewer could only find discussion of LOH/heterozygosity.*

   Response: In the Methods section, the authors state that the ploidy of the mutation-accumulation (MA) lines was assessed by flow cytometry both before and at the end of the experiment, specifically “to verify the ploidy level.” Moreover, the Discussion explicitly states: “Overall, the ploidy level of MA lines generated from wild strains is extremely stable.” These statements support the conclusion that the study assessed and reported ploidy-level stability, although the main focus of the paper was on loss of heterozygosity.

   • If this process is really the predominant mechanism of ploidy shifts and generation of aneuploids, then testing mutants that are missing a gene required for sporulation, endoreduplication or mating should block the process and greatly reduce the frequency of triploid vs tetraploid isolates.*

   Response: In principle, impairing sporulation or mating would directly eliminate processes that define the SEM sequence and would therefore provide limited mechanistic information beyond the experiments presented here. Furthermore, the molecular mechanism underlying postmeiotic endoreplication remains unknown, we therefore consider this experiment an interesting avenue for future work, but beyond the scope of the present study.

   ---

   Significance* This paper has the potential to change ideas about the sources of polyploidy and aneuploidy in S. cerevisiae and to highlight intra-ascus events that drive some of it. The text and figures need to be clearer, and the storyline needs to be made more accessible to readers who are not super-specialized in this area(as indicated in specific comments above). A few tables or figure that compare more assumptions and expectations of the two major models that are compared (and perhaps note other possible models as well) would go a long way to underpinning the data and making the case for the claims (once they are appropriately tempered based on the data). As currently written, this would reach a specialized audience interested in mechanisms of generating ploidy changes in the model yeast. It would be much better if written for a broader audience including those working on ploidy issues in many different fungi.*

   Response: We thank Reviewer 1 for this positive assessment and for emphasizing the broader significance of our study. We agree that the manuscript needed to be more accessible to readers beyond the yeast-genetics community. In response, we have substantially revised the text and figures and clarified the experimental logic and terminology. We believe that these changes have improved the clarity, accessibility, and overall impact of the manuscript.

   Reviewer #2

   Evidence, reproducibility and clarity

   This study demonstrates that ploidy can increase one level at a time with a reduced gamete that undergoes endoreplication mating with a typical sibling gamete, highlighting a bigger role for triploids as an intermediate step to polyploidized lineages. It is by endoreplication, not whole genome duplication, resolving an existing uncertainty. It also resolves a major question of yeast triploid incidence as likely being from this mechanism and not partial tetraploid reversion (this latter point could be better emphasized in the abstract). The figures need a little more layman's explanation of the allele balance ratio data. The methods need a little more detail for reproducibility.

   Response: We thank the reviewer for his very positive assessment and recommendation for publication. In the revised abstract, we now explicitly state that triploids are key intermediates in stepwise polyploidization, rather than merely transient products of tetraploid regression toward diploidy. In the Result section, we have expanded the description of the allele balance ratio (ABR) concept, which is used in several figures. Finally, we have added an Extended Methods section providing additional experimental details on cytometric analyses, mating and zygote formation, creeping assays, and the calculation of mating scores.

   Minor: I give small points of improvement through the manuscript but see no major issues and recommend publication otherwise.

   Line 24: Mention the other autopolyploidization mechanisms

   Response: We added a sentence in the abstract to say that polyploidization is generally attributed to mitotic genome doubling or unreduced gamete fusion.

   Line 30: Mention the triploid bridge from plant studies

   Response: We added a sentence in the abstract to say that triploids are central intermediates in stepwise polyploidization, analogous to the triploid bridge described in plants.

   Line 50: Worth a mention there that ohnogenes (gene copies from WGD) have a higher incidence of becoming oncogenes, see work of Hevre Isambert.

   Response: We thank the reviewer for pointing us to this reference. We have added a sentence noting that, in vertebrates, genes retained from ancient whole genome duplication events, known as ohnologs, are disproportionately enriched among genes implicated in cancer and dominant genetic disorders.

   Line 65: one sentence needed on what the canonical lifestyle of yeasts is

   Response: We have added a description of the canonical life cycle, explaining that diploid cells carry the two mating types, MATa and MATα, and normally undergo meiosis and sporulation to produce four haploid spores, two MATa and two MATα, which remain enclosed within an ascus. Upon germination, spores of opposite mating types can mate with one another, thereby restoring the diploid state.

   Line 75: Some general statement needed about any interesting or unique population level functions of the triploids and/or tetraploids

   Response: We found it difficult to add a general statement at this point without interrupting the flow of the narrative. Moreover, at the beginning of the Introduction, we already state that polyploidy promotes genomic and phenotypic diversification, facilitates adaptation to stress, and contributes to ecological success and domestication. Nevertheless, we have clarified the composition of the “natural polyploid” category, which here includes both wild isolates recovered from natural niches and domesticated isolates originating from industrial or other anthropogenic environments. We also clearly state in the discussion section that polyploid yeast strains are widely used in fermentation industries, including brewing and baking, and are also frequently isolated from clinical settings.

   Line 82: I have no problem with this SEM acronym and think the sporulation/endoreplication/mate description accurate, but perhaps authors might want to reconsider for search optimization of their results because SEM also stands for scanning electron microscopy

   Response: We understand the reviewer’s concern regarding the use of “SEM,” which is also commonly used to refer to scanning electron microscopy. However, we have not identified a more suitable acronym, and we feel that SEM accurately and concisely represents the Sporulate–Endoreplicate–Mate mechanism proposed here. We therefore prefer to retain the acronym, while recognizing that it is also used in other contexts.

   Figure 1C: It's not clear what changes from step 2 to 3, and that it shows mating. It looks like a selfing process. Color labelling of spores would make this idea clearer.

   Response: We assume that the reviewer is referring to Fig. 1B, as there is no Fig. 1C. We thank the reviewer for drawing our attention to this lack of clarity. We have clarified that step 2 corresponds to the endoreplication of one spore, whereas step 3 corresponds to mating between the endoreplicated spore and one intact spore. We have also modified the illustrations of the endoreplication event and the mating bridge to make these steps more apparent.

   Figure 2A: Can you put this same color scheme into figure 1C

   Response: We assume that the reviewer is referring to Fig. 1B, as there is no Fig. 1C. We prefer to retain the gray shading in this figure to distinguish the number of chromosome sets in vegetative 2n and gametic 1n cells from overall ploidy. Throughout the figures, ploidy is consistently represented using green, orange, and brown for 2x, 3x, and 4x cells, respectively. Although this distinction may be subtle, we consider it important because the SEM mechanism can operate across different ploidy levels.

   Figure 2B: What is the unique dot in the middle of the light field of the ATR strain?

   Response: The dot represents the global mating score, calculated as the average of the mating, zygote formation, and creeping assay scores. This score was calculated for all strains but was not always visible because it lay at the extremity of the corresponding bars. To simplify the panel and avoid potential confusion, we have removed these dots from the figure.

   Line 184: Explain what a creeping phenotype is earlier on in the main manuscript and relevance to ploidy variation

   Response: We have clarified the meaning and relevance of the creeping phenotype at its first occurrence. In the Fig. 2 legend, we now define it as “a mating-associated aggregation phenotype used as an additional readout of mating competence.”

   Line 209: do you want to definitively attribute a relative contribution in concrete numbers to either mutational load or chromosome mis-segregation, e.g. mutational load is much worse (the ~5% viability versus the 50% viability)?

   Response: We agree that the comparison is informative, but we do not think that the relative contributions of mutational load and chromosome mis-segregation can be quantified directly, because the natural triploids examined here have different genetic backgrounds. We have therefore revised the text to place greater emphasis on the likely contribution of mutational load while avoiding an unsupported numerical attribution: “The substantially lower spore viability of natural triploids (5.0–7.8%) suggests that, in addition to triploidy-associated segregation defects, meiotic segregation unmasks a major contribution from recessive deleterious variants in their genomes, including several heterozygous high-impact variants in essential genes.”

   Figure 2: please put keys in the figures instead of having them written out in the captions

   Response: We have added the relevant keys directly to the figures, as requested, rather than describing them only in the captions.

   Figure 3C: I do not understand the y-axis of the ABR figure or the data distribution patterns and how they demonstrate higher ABR complexity/correlation to the colored copies figure above it. Can you explain better in the caption how to perceive this pattern? (That is, I get the idea that it about ratios to a reference allele to interpret ploidy, but it is a struggle for a lay person to interpret what look like different patterns (e.g .a split cloud over 0.75 and 0.25, and a concentrated cloud at 0.5) could yield the same ploidy assignment (4x) and how that can be differentiated from the other ploidy levels.

   Response: We have added the following explanation to the legend of Fig. 3C: “In the chromosome-level ABR profile, each gray point represents a heterozygous variant along the chromosome. ABR profiles show the fraction of sequencing reads carrying the alternative allele at heterozygous sites. Only variants with ABR values between 0.125 and 0.875 are shown (see Methods).” We have also added an explanation in the main text: “Third, allele balance ratios (ABRs), which measure the relative abundance of alternative alleles at heterozygous sites, showed patterns consistent with genome doubling (Fig. 3C). In four-copy chromosomes, heterozygous variants had an ABR of ~ 0.5, as expected when two-copy chromosomes are duplicated (Fig. 3D). By contrast, two-copy chromosomes lacked heterozygous variants, consistent with duplication of chromosomes that were initially present in a single copy (Fig. 3C and 3D; Supplementary Fig. 3).”

   Figure 4E: Same comment as above.

   Response: We have added the following explanation to the legend of Fig.4E: “ABR profiles show the fraction of sequencing reads carrying the alternative allele at heterozygous sites (y-axis), with each gray point representing a variant along the chromosome (x-axis). In ALH-3x, ABRs cluster around ~0.33 and ~0.67 on three-copy chromosomes, corresponding to 1:2 and 2:1 allelic ratios, respectively. In ALH-4x, ABRs cluster around ~0.25, ~0.5, and ~0.75 on four-copy chromosomes, corresponding to 1:3, 2:2, and 3:1 allelic ratios, respectively.”.

   Line 490: Intriguing idea about application, but can you be more specific? In plant (and some animal) breeding, polyploidy is an important tool for creating larger/more-stress resistant/ purposefully sterile strains (to prevent invasiveness in natural environments). Can you suggest a real benefit to a polyploidized yeast strain for baking and brewing?

   Response: We have not modified the text, as we do not yet have sufficient evidence to propose a specific, validated benefit of SEM-generated polyploid strains for baking or brewing. We already note that polyploid yeast strains are widely used in fermentation industries, including brewing and baking, and are also frequently isolated from clinical settings. Although SEM could potentially provide a route to generate novel polyploid strains with desirable industrial traits, such as improved fermentation performance or stress tolerance, whether this mechanism can be reliably harnessed for strain improvement remains unknown.

   Line 249: Very nice set of experiments proving viable aneuploids

   Response: We thank the reviewer for this nice comment.

   Line 518: I know the protocol is referenced, but a few more details here on the flow cytometry would be helpful. What was the buffer used, what was the flow cytometer, what was the number of events.

   Response: We have added all details in an Extended method section and also added few more details in the main Methods section.

   Line 573-574: Same comment, just some brief details to give a general idea.

   Response: we have added a few more details.

   Line 569-571: Something off here because a flow cytometry step is described twice. What is the distinction between the first description and the second description?

   Response: We thank the reviewer for pointing out this ambiguity. We used the same flow-cytometry readout for ploidy determination but two different sample-preparation procedures depending on the purpose of the analysis. A rapid preparation was used for initial screening of ascus-derived colonies, whereas a more thorough preparation, including PBS washing, was used for subsequent ploidy validation. We have revised the Flow cytometry and Ascus micromanipulation sections to make this distinction explicit.

   Line 617: Correct various formatting errors in references: italicize species, do not use all caps or capitalize all words in titles, etc.

   Response: We have corrected the formatting errors in the references. Species names are now italicized, and titles have been standardized by removing unnecessary capitalization and all-capital formatting.

   Supplementary Figure 1 caption: The data points falling along a gradient value instead of discrete integers indicates aneuploidy and repeat measures. Please mention both in the caption, and add standard errors if possible.

   Response: We thank the reviewer for pointing out that the representation was not sufficiently clear. The data points in Supplementary Fig. 1 do not represent repeated measurements. Triangles correspond to individual monosporic isolates derived from diploid strains, whereas circles correspond to the progeny obtained after crossing these isolates with compatible haploid testers. Therefore, standard errors are not applicable. We have revised the figure legend to make this distinction explicit. We also clarify that the non-integer ploidy values represent quantitative DNA-content estimates and can reflect deviations from complete euploid chromosome sets.

   Supplementary captions: The ABR patterns are much better explained here than in the main figure captions. Please also put a form of these descriptions in the main figure captions, even if it seems repetitive.

   Response: We have updated the main figure legends to include the clearer, more detailed descriptions of ABR patterns that were previously only in the supplementary captions, ensuring consistent and self-contained explanations across all ABR panels.

   * *

   Significance

   *I am a general polyploid evolution, not yeast specialist, so apologies if I have gaps in reviewing the yeast-specific aspects. But from my view this is a well done and well written study that establishes the mechanism by which yeast can be polyploidized. This is in reference the important evolutionary phenomenon of whole genome duplication conferring adaptive advantages in many taxa, but as authors touch on at the end, breeding improvement. The novelty is slightly inaccurately presented in the abstract (more accurately presented in the actual text). As authors note, there are many studies demonstrating the importance of the triploid bridge, in plant allopolyploids mostly. This should be brought into some comparative discussion. Autopolyplodizaton happening through the usual routes of lack of gametic reduction, etc. as the authors indicate in introduction, should be touched upon in abstract.

   It is a stretch to say this is a "new framework for polyploid genome formation across eukaryotes" but it is certainly a good work perhaps widely applicable for fungi, which are understudied for polyploid evolution. This theme could be expanded on throughout the work for a more specific but more intriguing direction. For example what is the ecological significance for such high triploid/tetraploid incidence in yeast/fungi? What could the accessibility of this polyploidizaton mechanism imply for their evolution?*

   Response: We thank the Reviewer for this supportive assessment of our study. We are pleased that the work was recognized as well conducted and as establishing a mechanism underlying an important evolutionary phenomenon with broad implications for adaptation and genome evolution.

   Reviewer #3

   Evidence, reproducibility and clarity:

   In this manuscript, titled "A postmeiotic route to stepwise polyploidy", Fischer and co-workers describe a novel model for the development of polyploidy in yeast, and hence in possibly many other organisms. Based on careful examination of individual examples and of a broad panel of natural isolates, as well as experimental reconstitution they describe a scenario whereby right after meiosis a germinating spore undergoes endoreplication, hence becoming diploid, before it mates with a neighboring spore, possibly a meiotic sibling, to generate a triploid. This sequence, sporulation, endoreplication followed by mating, is abbreviated SEM. Subsequent reiteration of this process leads ultimately to the generation of tetraploids or near tetraploids. Supporting this model, the frequency of triploids, of heterozygosity and of aneuploidy in natural isolates is much higher than predicted by the more classic model whereby tetraploids are generated by division failure or endoreplication of a diploid. This scenario is also shown to be prevalent in the case of heterothallic and much less frequent in homothallic strains, supporting the notion that endoreplication and mating are early events after spore germination. This scenario also suggests that triploids are much better able to generate viable spores that themselves engage in mating than generally anticipated.

   The data is solid, based on a solid dataset of natural isolates and experimental derivatives from them, including full genome sequences. The argumentation based on the data presented is very convincing.

   Response: We sincerely thank the reviewer for this positive assessment of our manuscript.

   A few presentation points would, however, benefit from further attention by the authors before publication.

   1- The notion of endoreplication will suggest a very precise scenario to many cell biologists, namely the succession of two S-phases without mitosis, and may appear as excluding other scenario with the same output, such as abortive mitosis or failed cytokinesis. It would therefore be useful for the authors to perhaps find another wording or to clearly define what they understand under endoreplication. Indeed, the data as it stands does not allow distinguishing between these diverse possibilities and the discussion itself demonstrates that the authors are open to a variety of possible mechanisms. This discussion could also benefit from being a bit better substantiated, to ensure that there is no misunderstanding.

   Response: We fully agree that we cannot conclude on the molecular mechanism involved in the endoreplication step and we acknowledge this present limitation. However, we are using the word endoreplication on purpose to include endocycling, abortive mitosis and cytokinesis failure as it is described in the review by D. T. Fox, R. J. Duronio (2013). We also clearly state this at the beginning of the result section: “The molecular mechanism underlying this genome doubling event is currently unknown. Here, we use endoreplication as a general term for genome doubling that may result either from two successive rounds of DNA replication without mitosis or from an abortive mitosis or failed cytokinesis (30).” So, we believe that the message is clearly conveyed and that we don’t need to find another wording.

   2- The authors report that the SEM sequence is well identifiable in heterothallic isolates and barely so in homothallic ones. However, they do not discuss why they think this is. Upon spore germination, recent data have indicated that cells needs to undergo several rounds of divisions before becoming able to mate (PMID: 30355051). This should give time to both homothallic and heterothallic spores to "endoreplicate". Therefore, why SEM is dependent on heterothallism is not intuitive and would profit of some discussion.

   Response: We thank the reviewer for raising this important question. We have added a discussion of the difference between heterothallic and homothallic isolates and now propose two possible, although speculative, explanations:

   “By contrast, spores able to self-mate may rapidly restore diploidy through mating and therefore be less likely to enter the SEM pathway. Alternatively, loss of mating-type switching itself could influence endoreplication: if cells attempt to initiate mating-type switching but the process is impaired, this failure could trigger a cellular response that favors genome duplication. Whether either of these mechanisms, or another mechanism, promotes entry into the SEM pathway remains to be determined.”

   This discussion makes clear that the proposed explanations are hypotheses and that the molecular basis of the apparent association between heterothallism and SEM remains unresolved.

   3- The figures are not particularly intuitive and it takes some time to understand what data is represented and how. For examples, this reviewer found figure 2B quite cryptic at first. It would be useful to provide a schematic of the different tests used and of how the mating score is derived from them. Figure 3B refers to BBT, BAD and CRE. It would be useful to remind the readers in the legend what these abbreviations stand for, without having to search for them in the supplementary material.

   Response: We thank the reviewer for drawing our attention to this issue. We have made substantial efforts to improve the clarity and accessibility of all figures, not only Fig. 2B. In particular, we have added schematics illustrating the different assays and how the global mating score is calculated. We have also revised the figure legends extensively and now define abbreviations such as BBT, BAD, and CRE directly in the relevant legends, so that the necessary information is accessible without consulting the Supplementary Material.

   **Referees cross-commenting**

   I agree with the two other reviewers that the text is written for a highly specialised audience. As a yeast geneticist, this did not disturb me too much, but it was even for me somewhat jargonous at places. The authors would strongly increase the impact of their paper if they would make it more approachable for a broad audience.

   I also agree that bringing a broader comparison with the data available would help judging better about the prevalence of the SEM scenario.

   Response: We sincerely thank all three reviewers for drawing our attention to this issue. As yeast geneticists, we recognize that concepts familiar to us may be specialized and not readily accessible to researchers from other fields. We have therefore made substantial efforts to rewrite the manuscript and make it accessible to a broad audience. We simplified several concepts and replaced specialized terminology with more explicit wording; for example, “homothallic” and “heterothallic” have been replaced throughout by “self-mating” and “non-self-mating” strains. We also substantially improved the figures so that the main results can be understood more intuitively without relying exclusively on the detailed text. Finally, we placed the SEM mechanism in a broader context by discussing it alongside other established mechanisms of polyploidization. Overall, we believe that these revisions have significantly improved the clarity, accessibility, and impact of the manuscript.

   Significance:

   Overall, this manuscript proposes a very original model that departs from classically accepted scenarios in interesting manners. It is also very well supported by a large body of observations and careful analyses. This study, therefore, renews our views and concepts about genome evolution, and the emergence of polyploid phenotypes. Whereas I expect that this work might steer quite some debate, I find it a well-grounded and very useful addition to the discussion. As such, this manuscript will address a broad audience in the fields of evolution, genome structure, molecular biology, cell biology and biotechnology.

   Response: We are pleased that the Reviewer considers our work a potentially important contribution to current models of genome evolution and polyploidization. We share the view that the discovery of SEM could represent a conceptual advance in our understanding of yeast genome evolution and may also have broader implications for polyploidization beyond yeast.

   Read the original source
2. 

   Review Commons

   Sep 24, 2026

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #3

   Evidence, reproducibility and clarity

   In this manuscript, titled "A postmeiotic route to stepwise polyploidy", Fischer and co-workers describe a novel model for the development of polyploidy in yeast, and hence in possibly many other organisms. Based on careful examination of individual examples and of a broad panel of natural isolates, as well as experimental reconstitution they describe a scenario whereby right after meiosis a germinating spore undergoes endoreplication, hence becoming diploid, before it mates with a neighboring spore, possibly a meiotic sibling, to generate a triploid. This sequence, sporulation, endoreplication followed by mating, is abbreviated …

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #3

   Evidence, reproducibility and clarity

   In this manuscript, titled "A postmeiotic route to stepwise polyploidy", Fischer and co-workers describe a novel model for the development of polyploidy in yeast, and hence in possibly many other organisms. Based on careful examination of individual examples and of a broad panel of natural isolates, as well as experimental reconstitution they describe a scenario whereby right after meiosis a germinating spore undergoes endoreplication, hence becoming diploid, before it mates with a neighboring spore, possibly a meiotic sibling, to generate a triploid. This sequence, sporulation, endoreplication followed by mating, is abbreviated SEM. Subsequent reiteration of this process leads ultimately to the generation of tetraploids or near tetraploids. Supporting this model, the frequency of triploids, of heterozygosity and of aneuploidy in natural isolates is much higher than predicted by the more classic model whereby tetraploids are generated by division failure or endoreplication of a diploid. This scenario is also shown to be prevalent in the case of heterothallic and much less frequent in homothallic strains, supporting the notion that endoreplication and mating are early events after spore germination. This scenario also suggests that triploids are much better able to generate viable spores that themselves engage in mating than generally anticipated.

   The data is solid, based on a solid dataset of natural isolates and experimental derivatives from them, including full genome sequences. The argumentation based on the data presented is very convincing.

   A few presentation points would, however, benefit from further attention by the authors before publication.

   1- The notion of endoreplication will suggest a very precise scenario to many cell biologists, namely the succession of two S-phases without mitosis, and may appear as excluding other scenario with the same output, such as abortive mitosis or failed cytokinesis. It would therefore be useful for the authors to perhaps find another wording or to clearly define what they understand under endoreplication. Indeed, the data as it stands does not allow distinguishing between these diverse possibilities and the discussion itself demonstrates that the authors are open to a variety of possible mechanisms. This discussion could also benefit from being a bit better substantiated, to ensure that there is no misunderstanding.

   2- The authors report that the SEM sequence is well identifiable in heterothallic isolates and barely so in homothallic ones. However, they do not discuss why they think this is. Upon spore germination, recent data have indicated that cells needs to undergo several rounds of divisions before becoming able to mate (PMID: 30355051). This should give time to both homothallic and heterothallic spores to "endoreplicate". Therefore, why SEM is dependent on heterothallism is not intuitive and would profit of some discussion.

   3- The figures are not particularly intuitive and it takes some time to understand what data is represented and how. For examples, this reviewer found figure 2B quite cryptic at first. It would be useful to provide a schematic of the different tests used and of how the mating score is derived from them. Figure 3B refers to BBT, BAD and CRE. It would be useful to remind the readers in the legend what these abbreviations stand for, without having to search for them in the supplementary material.

   Referees cross-commenting

   I agree with the two other reviewers that the text is written for a highly specialised audience. As a yeast geneticist, this did not disturb me too much, but it was even for me somewhat jargonous at places. The authors would strongly increase the impact of their paper if they would make it more approachable for a broad audience. I also agree that bringing a broader comparison with the data available would help judging better about the prevalence of the SEM scenario.

   Significance

   Overall, this manuscript proposes a very original model that departs from classically accepted scenarios in interesting manners. It is also very well supported by a large body of observations and careful analyses. This study, therefore, renews our views and concepts about genome evolution, and the emergence of polyploid phenotypes. Whereas I expect that this work might steer quite some debate, I find it a well-grounded and very useful addition to the discussion. As such, this manuscript will address a broad audience in the fields of evolution, genome structure, molecular biology, cell biology and biotechnology.

   Read the original source
3. 

   Review Commons

   Sep 24, 2026

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #2

   Evidence, reproducibility and clarity

   This study demonstrates that ploidy can increase one level at a time with a reduced gamete that undergoes endoreplication mating with a typical sibling gamete, highlighting a bigger role for triploids as an intermediate step to polyploidized lineages. It is by endoreplication, not whole genome duplication, resolving an existing uncertainty. It also resolves a major question of yeast triploid incidence as likely being from this mechanism and not partial tetraploid reversion (this latter point could be better emphasized in the abstract). The figures need a little more layman's explanation of the allele balance ratio data. The methods …

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #2

   Evidence, reproducibility and clarity

   This study demonstrates that ploidy can increase one level at a time with a reduced gamete that undergoes endoreplication mating with a typical sibling gamete, highlighting a bigger role for triploids as an intermediate step to polyploidized lineages. It is by endoreplication, not whole genome duplication, resolving an existing uncertainty. It also resolves a major question of yeast triploid incidence as likely being from this mechanism and not partial tetraploid reversion (this latter point could be better emphasized in the abstract). The figures need a little more layman's explanation of the allele balance ratio data. The methods need a little more detail for reproducibility.

   Minor:

   I give small points of improvement through the manuscript but see no major issues and recommend publication otherwise.

   • Line 24: Mention the other autopolyploidization mechanisms

   • Line 30: Mention the triploid bridge from plant studies

   • Line 50: Worth a mention there that ohnogenes (gene copies from WGD) have a higher incidence of becoming oncogenes, see work of Hevre Isambert.

   • Line 65: one sentence needed on what the canonical lifestyle of yeasts is

   • Line 75: Some general statement needed about any interesting or unique population level functions of the triploids and/or tetraploids

   • Line 82: I have no problem with this SEM acronym and think the sporulation/endoreplication/mate description accurate, but perhaps authors might want to reconsider for search optimization of their results because SEM also stands for scanning electron microscopy

   • Figure 1C: It's not clear what changes from step 2 to 3, and that it shows mating. It looks like a selfing process. Color labelling of spores would make this idea clearer.

   • Figure 2A: Can you put this same color scheme into figure 1C

   • Figure 2B: What is the unique dot in the middle of the light field of the ATR strain?

   • Line 184: Explain what a creeping phenotype is earlier on in the main manuscript and relevance to ploidy variation

   • Line 209: do you want to definitively attribute a relative contribution in concrete numbers to either mutational load or chromosome mis-segregation, e.g. mutational load is much worse (the ~5% viability versus the 50% viability)?

   • Figure 2: please put keys in the figures instead of having them written out in the captions

   • Figure 3C: I do not understand the y-axis of the ABR figure or the data distribution patterns and how they demonstrate higher ABR complexity/correlation to the colored copies figure above it. Can you explain better in the caption how to perceive this pattern? (That is, I get the idea that it about ratios to a reference allele to interpret ploidy, but it is a struggle for a lay person to interpret what look like different patterns (e.g .a split cloud over 0.75 and 0.25, and a concentrated cloud at 0.5) could yield the same ploidy assignment (4x) and how that can be differentiated from the other ploidy levels.

   • Figure 4E: Same comment as above.

   • Line 490: Intriguing idea about application, but can you be more specific? In plant (and some animal) breeding, polyploidy is an important tool for creating larger/more-stress resistant/ purposefully sterile strains (to prevent invasiveness in natural environments). Can you suggest a real benefit to a polyploidized yeast strain for baking and brewing?

   • Line 249: Very nice set of experiments proving viable aneuploids

   • Line 518: I know the protocol is referenced, but a few more details here on the flow cytometry would be helpful. What was the buffer used, what was the flow cytometer, what was the number of events.

   • Line 573-574: Same comment, just some brief details to give a general idea.

   • Line 569-571: Something off here because a flow cytometry step is described twice. What is the distinction between the first description and the second description?

   • Line 617: Correct various formatting errors in references: italicize species, do not use all caps or capitalize all words in titles, etc.

   • Supplementary Figure 1 caption: The data points falling along a gradient value instead of discrete integers indicates aneuploidy and repeat measures. Please mention both in the caption, and add standard errors if possible.

   • Supplementary captions: The ABR patterns are much better explained here than in the main figure captions. Please also put a form of these descriptions in the main figure captions, even if it seems repetitive.

   Referees cross-commenting

   I appreciate that the other two reviewers appear to be yeast experts and filled in my gaps of knowledge in this system. I think point about clarifying interpretation around the data variance very important. It is fine if the ploidy increase events are not common or uneven across strains, but it needs to be contextualized in big picture biological relevance. If data also does not show the degree of ploidy increase as described in the main text, the latter also needs to be adjusted.

   Significance

   I am a general polyploid evolution, not yeast specialist, so apologies if I have gaps in reviewing the yeast-specific aspects. But from my view this is a well done and well written study that establishes the mechanism by which yeast can be polyploidized. This is in reference the important evolutionary phenomenon of whole genome duplication conferring adaptive advantages in many taxa, but as authors touch on at the end, breeding improvement. The novelty is slightly inaccurately presented in the abstract (more accurately presented in the actual text). As authors note, there are many studies demonstrating the importance of the triploid bridge, in plant allopolyploids mostly. This should be brought into some comparative discussion. Autopolyplodizaton happening through the usual routes of lack of gametic reduction, etc. as the authors indicate in introduction, should be touched upon in abstract.

   It is a stretch to say this is a "new framework for polyploid genome formation across eukaryotes" but it is certainly a good work perhaps widely applicable for fungi, which are understudied for polyploid evolution. This theme could be expanded on throughout the work for a more specific but more intriguing direction. For example what is the ecological significance for such high triploid/tetraploid incidence in yeast/fungi? What could the accessibility of this polyploidizaton mechanism imply for their evolution?

   Read the original source
4. 

   Review Commons

   Sep 24, 2026

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #1

   Evidence, reproducibility and clarity

   Summary

   This paper analyzes the process of ploidy alterations amongst a set of diverse yeast (S. cerevisiae) strains and documents the sequential appearance of ~diploid, ~ triploid and ~tetraploid isolates from asci. It argues that this is the predominant mechanism of ploidy increases and that it explains many findings in other studies.

   Major comments

   1. The key conclusion of the paper is that the series of consecutive events described is "the predominant natural route to polyploidy in yeast". However, the events shown remain relatively rare (and if they are not rare, no comparison to frequency of other events is provided). In …

   Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

   Learn more at Review Commons

   ---

   Referee #1

   Evidence, reproducibility and clarity

   Summary

   This paper analyzes the process of ploidy alterations amongst a set of diverse yeast (S. cerevisiae) strains and documents the sequential appearance of ~diploid, ~ triploid and ~tetraploid isolates from asci. It argues that this is the predominant mechanism of ploidy increases and that it explains many findings in other studies.

   Major comments

   1. The key conclusion of the paper is that the series of consecutive events described is "the predominant natural route to polyploidy in yeast". However, the events shown remain relatively rare (and if they are not rare, no comparison to frequency of other events is provided). In addition, the experiments start with a selection of strains picked for their ability to mate.

   a) A figure illustrating the frequency of each step in the process would help put the results into larger context.

   1. While the data show that the series of sporulation/endoreplication/and mating appears to generate cells with altered ploidy and extensive aneuploidy, the evidence that this is "the predominant natural route" is difficult to extract from the work as presented. There are references to prior papers, but the key points in those papers should be presented in a manner that is convincing-perhaps in a table that compares issues addressed in this work and that build the argument made in favor of this SEM mechanism being 'predominant (e.g, frequency of aneuploidy, frequency of triploidy, tetraploidy, ploidy stability etc.)

   a) An example of this is the statement in lines 62-64 that "8-20% of natural isolates are autopolyploid and triploids are at least as frequent as tetraploids" , but Peter et al found <5% triploids among 1011 yeasts and a similar number of tetraploids. Providing the actual data being quoted across the literature would help make this argument more specifically.

   1. Figure 2B is extremely confusing and would benefit by making it easier to understand.

   a) First, it would help to understand how these particular strains were chosen from the 728 starting strains. The text says that it chose those that could mate, but if step 1 is sporulation, why not test the hundreds of strains that could sporulate (based on Table S2). The left part of Fig 2B illustrates a mating score inferred based on 3 assays. Are the a or alpha designations and listing of hetero vs homo zygosity from before or after the experiment? How part C relates to part B of the figure is unclear, one would need to trace the strains from the left of 2B to the bottom of 2C.

   b) The title of the legend states that "endoreduplication during spore germination enables diploid to triploid transition" (line 180). Yet, the 'triploids generated are primarily <3N (presumably highly aneuploid) and a good proportion of the 'triploid' strains are much closer to diploid ploidy levels. Finally, only 2/23 of the homothallic isolates (those that can switch mating type and then mate with their progeny) display this 'triploid identity'. Please explain how this argues for a prevalent mechanism? And is it possible that the selection of specific strains for this experiment may have minimized the detection of endoreduplicaton (which does not require mating or sporulation if it is due to mitotic events)?

   c) In Fig 2C, there are a few red or blue spots in the homothallic strains, so the term 'exclusively' in line 190 cannot be true. It is also not clear why endoreduplication would expect to generate strains that are the same ploidy level as the parents, unless the intention is that endoreduplication AFTER meiosis is expected to give diploid progeny. Please be more explicit as there are many unstated assumptions here that are difficult to follow.

   1. The triploidy to tetraploidy step (starting at line 196) appears to be built upon analysis of 3 euploid triploid isolates (from the 1011 collection, not from experiments in Figure 2). These exhibit very low spore viability (presumably due to chromosome segregation defects known to be prevalent in spores from triploid parents). Accordingly, these meioses gave rise to aneuploid spores. It is argued that any strains with tetrasomic chromosomes must have arisen via endoreduplication. However, highly aneuploid isolates exhibit high levels of chromosome instability, such that missegregation of extra chromosomal pairs would need to be ruled out to reach this conclusion. Furthermore, if the '1.5x' spores that mate carried 1x of most chromosomes and 2x of other chromosomes (e.g., chromosomes 4,5,6) -what rules out the possibility that they mated with another spore that also had 1x of the same chromosomes and 2x of the same ones (e.g., chromosomes 4,5,6) .

   a) The illustration of this process in Figure 3 is confusing. For example, the text says that the 3X spore in the ascus after E2 is a/alpha/alpha, but the figure says it's only alpha/alpha. Does it have only alpha/alpha because it has only two copies of chr3 despite being ~3X? Assuming it has only MATalpha, it could then mate with a 1.5x spore that had either 1 or 2 copies of Chr3, and therefore was either a/a or alpha/alpha.

   b) In Fig. 3C, why is the ABR data for ChrV, VI, X, XI, XV and XVI so sparse?

   c) There is a focus on monosporic asci (i.e., only one spore survives the meiosis). Please explain the rationale for this and the assumptions that underlie it.

   1. Fig 4 looks at the whole ascus, rather than selected spores. And asks about their DNA content (4A), presumably after sporulation and then whatever mating or endoreduplication+ mating that goes on in the privacy of the ascus.... Most spores were haploid (didn't mate) or diploid (mated with ascus siblings) and 1-9% of the spores were triploid (with 2/8 analyzed having some tetrasomic and/or disomic chromosomes). While this is consistent with the model being proposed, can we rule out the possibility of a partial meiosis that yielded 2 1x spores and 1 2x spore and then mating between the 2x and 1x spores... There may be good arguments for this, but they were difficult to find.

   2. It apparently assumes that 3X spores arose only by endoreduplication and mating between sister spores.

   a) One confusing thing is that here it shows 2x and 3x progeny, but if we already know that triploid sporulation gives mostly ~1.5x spores (Figure 3) why are they labeled as 1x or 2x inside the asci in Figure 4B?

   1. The text is written for a highly specialized audience fluent in the fine points of S. cerevisiae mating, meiosis and ploidy change. Some simple explanations (e.g., homo vs heterothallism and how that is determined -presumably based on Ho gene function but not stated as such; 'bisexuality' - how does this differ from same sex mating and from mating of the same organism with its progeny that has switched to the opposite mating type?) What does dioecy mean for a yeast cell? These questions may seem naïve, but require extensive familiarity with the field; explaining the intended meaning would benefit allow a broader audience of readers to appreciate the fine points of this work.

   2. Arguments in the paragraph from line 388-398 are hard to follow. Some visual to bolster the explanation and help readers understand why stepwise is definitely more prevalent than saltation in all strains and conditions is needed.

   3. Lines 404-407-is this true for the natural tetraploids in the 1011 collection as well (that they lack 0.5 ABR peaks?) . Which stains are artificial vs natural vs from industrial fermentation sources.

   4. Paragraph 408-416-The argument is not easy to follow-if tetraploids are not so stable, then there should be lots of aneuploidy derived from them as well. Are there publications that have followed chromosome loss from triploid or tetraploids strains?

   5. While this paper may have established SEM as a mechanism of generating polyploidy and aneuploidy, it is difficult to understand if this mechanism is as 'predominant' as claimed. Please provide a figure or table to bolster this point. Furthermore, even if SEM is a frequent event, that doesnot rule out other mechanisms, the WGD being one of them. Thus, the authors should temper their statements to allow for other mechanisms that they show do also generate polyploids and aneuploids.

   6. Also, was the stepwise process recapitulated starting from haploids to the tetraploids for the same strains? My impression is that at the triploid stage more and 'better' triploids were chosen from the 1011 collection, so has the continuity of the process really been demonstrated definitively?

   1. Finally, how much does strain background play a role in the results and would attempting to repeat these experiments with different strains be likely to yield different results that would be simply ascribed to 'strain-specific effects'? The effort to look at many strains is admirable; explanations of why specific strains were chosen for the work over others would be helpful.

   Minor comments

   1. Are the strains selected for these studies truly 'wild' isolates or from domesticated yeasts? Could there be a difference between how domesticated/industrial yeasts and wild yeasts become polyploid? How do the strains study here compare with those from previous studies of polyploidization?

   2. One striking result that needs to be explained is why growth of a haploid lab strain (Ho deleted, one mating type) led to extremely rapid autodiploidization (mating type homozygous) that provides a growth advantage on glucose limiting medium (https://doi.org/10.1016/j.cell.2016.08.002). How frequently would that type of event happen when cells are stressed a bit?

   3. Lines 353-356-were sufficient numbers of tetraploids tested to make this a fair comparison?

   4. In figure 5C, what is the difference between an a mater, a mater and a bisexual mater and, related to this, what is dioecy in the context of yeast mating? It was previously described (ref 41) in polyploid fermentation strains. Can some context be provided on its role here?

   5. Lines 370-372-the order of the models in Fig1 A are saltation and then stepwise but are discussed here in the reverse order.

   6. Chromsome loss from tetraploids referred to from ref 30: Where in this paper is the stability of ploidy levels noted? In scanning all the figures, this reviewer could only find discussion of LOH/heterozygosity.

   7. If this process is really the predominant mechanism of ploidy shifts and generation of aneuploids, then testing mutants that are missing a gene required for sporulation, endoreduplication or mating should block the process and greatly reduce the frequency of triploid vs tetraploid isolates.

   Referees cross-commenting

   I agree with the other reviewers and think the manuscript will benefit from clarifications that make it much more accessible to general readers. I also think they need to be careful to avoid apparent overstatements that should be better substatiated.

   Significance

   Significance

   This paper has the potential to change ideas about the sources of polyploidy and aneuploidy in S. cerevisiae and to highlight intra-ascus events that drive some of it. The text and figures need to be clearer, and the storyline needs to be made more accessible to readers who are not super-specialized in this area(as indicated in specific comments above). A few tables or figure that compare more assumptions and expectations of the two major models that are compared (and perhaps note other possible models as well) would go a long way to underpinning the data and making the case for the claims (once they are appropriately tempered based on the data). As currently written, this would reach a specialized audience interested in mechanisms of generating ploidy changes in the model yeast. It would be much better if written for a broader audience including those working on ploidy issues in many different fungi.

   Expertise-I have extensive experience working with aneuploidy and, to some degree with polyploidy, but I am not involved directly in studies of S. cerevisiae meiosis and mating issues. Thus, I needed to do a lot of checking the referenced literature on those topics.

   Read the original source
5. 

   Version published to 10.64898/2026.02.20.706944 on bioRxiv

   Feb 20, 2026