SEVtras delineates small extracellular vesicles at droplet resolution from single-cell transcriptomes

Small extracellular vesicles (sEVs) are emerging as pivotal players in a wide range of physiological and pathological processes. However, a pressing challenge has been the lack of high-throughput techniques capable of unraveling the intricate heterogeneity of sEVs and decoding the underlying cellular behaviors governing sEV secretion. Here we leverage droplet-based single-cell RNA sequencing (scRNA-seq) and introduce an algorithm, SEVtras, to identify sEV-containing droplets and estimate the sEV secretion activity (ESAI) of individual cells. Through extensive validations on both simulated and real datasets, we demonstrate SEVtras’ efficacy in capturing sEV-containing droplets and characterizing the secretion activity of specific cell types. By applying SEVtras to four tumor scRNA-seq datasets, we further illustrate that the ESAI can serve as a potent indicator of tumor progression, particularly in the early stages. With the increasing importance and availability of scRNA-seq datasets, SEVtras holds promise in offering valuable extracellular insights into the cell heterogeneity.

minimal exosome counts per droplet required for accurate counting and classification, algorithm performance against complex background (other cell-free RNAs also encapsulated in the same droplets).4) For exosome source tracking, the authors made several assumptions (1) the transcriptional profile of an exosome is more similar to its cell of origin and (2) exosome release is affected by the biogenesis capacity of the original cell type.How generalizable are these assumptions across datasets?5) Since multiple vesicles are likely to be encapsulated within a single droplet, for droplets that contain heterogeneous mixture of exosomes derived from different cellular origins, how were they classified?6) How accurate is the approach as compared to other published tracking methods?Further investigation should be performed to assess the tracking accuracy, using pure vesicle populations derived from single cell source as well as known mixtures prepared with vesicles derived from different cells of origin.7) Induction and inhibition of exosome secretion could be included in multiple cell types to evaluate the responsiveness of ESAI.
Reviewer #2: Remarks to the Author: The authors present exoTras, a workflow that identifies exosomes containing droplets from single cell RNA-seq datasets.The algorithm is unique, novel and much needed to the field of extracellular vesicles (EVs).Though this reviewer is supportive of this article to be published, certain parts of the manuscript need to be updated.Each section needs more clarity as what was done and what is reported in the figures.Now it is quite vague and hard to understand.
1.The authors need to add additional data on the characterization of exosomes/EVs from single cell seq datasets and their workflow.What controls are in place to ensure that these are indeed EVs and not cell debris?Perhaps the authors can compare cells, debris (intentionally obtained with harsh treatment or stress), EVs for EV markers (CD63, CD81, Alix, Sytenin, TSG101) and negative (GRP96, Calnexin).This can be done on A549 cell dataset (Figure 1c) as well.The cells can be broken into debris by passing them through sieve or mechanical stress and compared with EVs and cell lysates.2. What quality control is placed to check that the sample used does not suffer from damage?Is there a possibility of including a negative control?Perhaps a RNA profile signature from debris? 3. ESAI is very interesting and informative.Though this is very useful, what if the EVs are migrating from adjacent cells?How does the algorithm tackle this?Hence the ESAI should also consider the cell type source for EVs and their locality in the tissue source of interest.4. Throughout the manuscript it is not clear what data is generated by the authors and those that are utilised from publicly available resources.Perhaps the authors can clarify in the manuscript.This also brings a point that the authors need to do this comprehensively on one dataset that they generated inhouse -where all characterization such as western blot for markers, quality control, debris all needs to be done.The existence of two exosomes group can also be shown comprehensively.The authors can apply exoTras to other datasets later like in the manuscript and shows its utility. 5. What does ESAI of 8.3% mean?Does it mean X of exosomes droplets / Y of total cells?How does the authors deal with EV subtypes and heterogeneity here?What about non-exosomes droplets but not cell droplets?The manuscript lacks clarity at various levels for every section.6. ESAI in figure 2e.How was this done?The legend is not clear?What does the error bar mean?Why is the error bar for Tumor adjacent not equal on both sides?7. The use of the term "exosome" need to be avoided throughput the manuscript including figures.Exosome refer to the RNA complex while exosomes (plural) refers to endocytic driven extracellular vesicles.8.The authors should use the term extracellular vesicles (EVs) or small EVs in general as opposed to exosomes.Exosomes are endocytically derived sEVs but no current technology can purify exosomes alone to 100% purity.
Reviewer #3: Remarks to the Author: The manuscript by He et al proposes a new algorithm ('exoTras') via which the RNA content and potential tissue origin of 'exosomes' (a subtype of extracellular vesicles) may be assessed.Hereto, the authors re-analyze droplet-based single cell RNAseq datasets.Droplet sequencing of (single) extracellular vesicles (EVs) isolated from culture medium or bodyfluids has been performed previously.However, these previous approaches did not allow tracing back the origin of the EVs to their secretory cell types.The new approach proposed here is based on the idea that libraries of droplet-based scRNAseq experiments are composed of cell-containing droplets, vacant droplets, cellular debris-containing droplets and exosome-containing droplets.It took me some time to understand the conceptual background of this approach, but I presume that it is based on the idea that droplets that contain RNA but no cells are thought to contain extracellular RNA (e.g.enclosed in EVs) released by the cells in question.I have doubts on the foundation of this approach based on the following arguments: 1.It is highly unlikely that the material that is found in non-cell-containing droplets of the scRNAseq experiments are actively released extracellular vesicles.The cells (either from cell cultures or detached from tissues by enzymatic treatments) are processed by washing and pelleting, and in some cases were subsequently cryopreserved.There was no recovery / culture step just prior to entering cells in the scRNAseq system (e.g.Chromium system).Therefore there has been no time/good condition for cells to actively release natural EVs.It is therefore far more likely that the extracellular RNAs are from rip-offs, which indeed would reflect the RNAome of the cells that had been processed.
In relation to the previous point, it is very difficult to substantiate that the observed differences in the amount of 'exosome' material between different (parts of) tissues is really indicative of differences in the amount of EVs these tissue cells release (indicated by the authors as the ESAI: 'exosome secretion activity index').The differences may just be a consequence of differences in the fragility of different cell types or the difficulty with which each cell type can be freed from the tissue environment.
2. The authors claim to be able to differentiate between 'debris' and 'exosomes'.First, it is currently not possible to differentiate between endosome-derived EVs (exosomes) and other types of actively released EVs once they are present outside cells.There is no evidence in this manuscript that the observed material is from genuine exosomes.Second, the adopted approach to differentiate between 'debris' and 'exosomes' is not sound.The reference EV transcriptome has been composed by manually curating EV-derived RNA gene sets from online available databases (exoCarta, ExRNA atlas).Especially the exoCarta database contains data of studies where only an enrichment step has been performed to isolate EVs.These samples are highly contaminated with non-EV-associated RNAs and therefore not suitable to use as specific reference for EV (or exosome)-associated RNAs.This, in combination with the previous point, severely complicates definitions of 'exosomes' versus 'debris'.
3. It is unclear how is assessed whether there is one or multiple EVs in a drop and the interpretation thereof (e.g. are these multiple EVs from one or multiple cell types?).
4. With regard to the previous arguments, the manuscript would have largely benefited from including solid experimental validation.Now, most of the 'validations' are simulations, which are not sufficient for substantiating the credibility of this entirely new approach.Experimental validation for assessing whether actively released EVs are measured and for the ESAI could for example be a droplet scRNAseq analysis of cultured cells treated or not with bafilomycin A1 as a reagent known to induce increased EV release.In addition, the capacity of exoTras to distinguish EV-versus debris-associated RNA could be validated by performing droplet scRNAseq of cells spiked with variable ratios of highly purified EVs and mechanically desintegrated cells (i.e.'debris').

Additional comment:
The authors should clarify in more detail what the application area of the ESAI and 'EV' typing is: do they expect that this exoTras method can be used for diagnosis of cancer, and which input material would be required for this?Is it expected that exoTras data reveal disease-associated exRNA transcriptomes that might be traced back in body fluids used as liquid biopsies?
Reviewer #4: Remarks to the Author: Exosomes are extracellular vesicles involved in the intercellular transportation of materials including nucleic acids, proteins, and/or other regulatory factors.They play vital roles in cell-cell communications regulating development, immune response, and disease onset and progression, and have been engineered as a vehicle for drug delivery.To elucidate the heterogeneity of exosome cargos, to identify their source of origin, and to unravel the regulatory roles of different types of exosomes have attracted much research.
The study by He et.al very nicely mounted to the ample availability of droplet-based single cell sequence data to develop an analytic tool set, exoTras, for exosome analysis.The main contribution includes a few algorithms to identify exosome containing droplets, to estimate exosome secretion activity for different types of cells, and to track back to their secretory cell types.The study also applied exoTras to analyze the droplet scRNA-seq data for normal tissues and cancer samples, and found exosome subtypes of different cargo type and cell source, as well as the elevated exosome secretion activity of aggressive epithelial cells in CRC and the potential usage of exosomes in biomarker for vascular invasion in early-stage cancer development.
This exoTras tool could be a powerful one that can dig into huge amount existing data from of dropletbased scRNA-seq studies (and requires no additional invest in new studies).The following are my comments: 1, it is not clear from the description how the authors deal with droplets of mixed content of cell debris and exosomes.I guess the authors know this, but the writing gives no clue on how these are handled and how it will affect the performance of exoTras.
2, the normalization of different samples is not well-described in the manuscript.How is the sequencing depth handled as a factor?I have a concern looking at Figure 1b--apparently total UMI affects the AUC.What about the other types of confounding factors considered?Some sample preparation protocols may deplete exosomes and the others may enrich them.Some experimental settings may favor a droplet to contain more exosomes and the others may disfavor it.The sequencing platform may favor the obtaining of specific exosome content.All of these factors may affect the sequencing results; it is very much like batch effects in single-cell data analysis.Without careful normalization and batch effect removal, it is dangerous to compare across samples.This is especially a factor that needs to be carefully taken care of for the cancer applications in page 7.
3, in the simulation study, the cell-exclusive sample are likely to be associated with exosomes as well.Please be precise.Also, will this be a problem?See above, what if the fraction of associated exosomes varies?Will this be a confounding factor?4, figure 1c, the correlation coefficients seem low even for droplets.This is rather expected given the little amount of material in each droplet and the data loss in scRNA-seq.Would exoTras benefit from data imputation?5, the study only investigated A549 cells.It would be very nice if the authors can perform more cell lines and to compare.It will also help the cell source tracing (for example, being a validation).
6, citation to supplementary figures should also be clear onto panels.
Minor 1, page 3, middle paragraph, I do not agree that the decrease of specificity is slight.
2, page 5~6, the data are from scRNA-seq studies of different tissues/organs, and the analyses are on constituent cell types.But the writing is a little confusing without the introduction of cell types.I made a mistake at the first time thinking that the authors confuse tissues and cell types.Better to mention the two to be less ambiguous.

Author Rebuttal to Initial comments Decision Letter, first revision:
Dear Fangqing, Thank you for submitting your revised manuscript "SEVtras delineates small extracellular vesicles at droplet resolution from single-cell transcriptomes" (NMETH-BC50796B).It has now been seen by the original referees and their comments are below.The reviewers find that the paper has improved in revision, and therefore we'll be happy in principle to publish it in Nature Methods, pending minor revisions to satisfy the referees' final requests and to comply with our editorial and formatting guidelines.
I strongly recommend addressing all the remaining concerns raised by Ref 3.
We are now performing detailed checks on your paper and will send you a checklist detailing our editorial and formatting requirements within two weeks or so.Please do not upload the final materials and make any revisions until you receive this additional information from us.
TRANSPARENT PEER REVIEW Nature Methods offers a transparent peer review option for new original research manuscripts submitted from 17th February 2021.We encourage increased transparency in peer review by publishing the reviewer comments, author rebuttal letters and editorial decision letters if the authors agree.Such peer review material is made available as a supplementary peer review file.Please state in the cover letter 'I wish to participate in transparent peer review' if you want to opt in, or 'I do not wish to participate in transparent peer review' if you don't.Failure to state your preference will result in delays in accepting your manuscript for publication.
Please note: we allow redactions to authors' rebuttal and reviewer comments in the interest of confidentiality.If you are concerned about the release of confidential data, please let us know specifically what information you would like to have removed.Please note that we cannot incorporate redactions for any other reasons.Reviewer names will be published in the peer review files if the reviewer signed the comments to authors, or if reviewers explicitly agree to release their name.For more information, please refer to our <a href="https://www.nature.com/documents/nr-transparentpeer-review.pdf"target="new">FAQ page</a>.ORCID IMPORTANT: Non-corresponding authors do not have to link their ORCIDs but are encouraged to do so.Please note that it will not be possible to add/modify ORCIDs at proof.Thus, please let your co-authors know that if they wish to have their ORCID added to the paper they must follow the procedure described in the following link prior to acceptance: https://www.springernature.com/gp/researchers/orcid/orcid-for-nature-researchThank you again for your interest in Nature Methods.Please do not hesitate to contact me if you have any questions.We will be in touch again soon.

Sincerely, Madhura
Madhura Mukhopadhyay, PhD Senior Editor Nature Methods Reviewer #2 (Remarks to the Author): The authors have addressed the concerns.
Reviewer #3 (Remarks to the Author): The revised manuscript by He et al describes the development and of an algorithm ('SEVtras') via which the RNA content and potential tissue origin of EVs may be assessed.Besides analysis of published droplet-based single cell RNAseq datasets, the authors also included experiments in which they performed sc-RNAseq experiments themselves to validate their conclusions.Compared to the original submission, this revised version has substantially improved.A large number of reviewer questions has been addressed.Yet, the rationale behind some of the experimental strategies is difficult to understand.In general, the manuscript also suffers from a large degree of complexity, lack of described experimental details, and figures being scattered over 3 different locations (main, extended data sets, and suppl figures).

Main comments:
General: Several of the figure panels (especially those in suppl and extended data figures) are not described in the main text.On top of that, there is continues switching between main figures, extended data figures and supplementals.All of this makes it extremely difficult to follow and understand the manuscript.
Feasibility of identifying sEV-containing droplets from scRNA-seq: -From a biology-perspective it is questionable whether cells that are being processed for sc-RNAseq analysis have sufficient time and are in good condition to actively release EVs as they would do in an in vivo or in vitro culture condition.Although this is a difficult question to answer, it is important to report exactly each step from harvesting of the tissue to dissociation of cells, to processing for sc-RNAseq analysis.I cannot find ANY information about this in the manuscript!-In Extended Data fig 1a-b the authors process tissues and look at EVs, but it is unclear how these experiments have been performed (no descriptions in Methods section!) and which question this answers.Are these the EVs that are present in the tissue prior to making single cell suspensions?If cells are washed during these procedures, all EVs present in tissues at the time of isolation will be discarded….How is this data relevant?-The authors aim to address the question above in Extended Data fig 1 c-e, but instead of assessing the processing steps to isolate the cells, the authors assessed whether the processing steps affected recovery of spiked-in EVs.That is not relevant, because the EVs present in the tissue at the time of harvesting are not the same as those that are released by cells at the timepoint of sc-RNAseq analysis.
-The methodology used in extended data fig 1 c-e is expression of nano-luc in cells and analyzing nanoluc signals in supernatants.However, this nano-luc can be released by cells in EV-associated and non-EVassociated forms, and is therefore not indicative for EV release.To prevent this, the nano-luc should be coupled to proteins frequently sorted into EVs (eg tetraspanins, as published before by several groups).

Performance evaluation of SEVtras:
In figure 1b-c the authors included wet lab experiments to validate their technique.Yet, the selected approaches raise several questions.
-The monensin treatment protocol should be added to the Methods section -The Methods section should also contain a detailed section on how the sc-RNAseq experiments were performed: how were the cells harvested, how many cells were processed, which washing steps were performed, how were libraries prepared, etc.
-The Methods section should also report to which number of MSC the EVs from 1e4 MSC were added to do the spike-in sc-RNAseq evaluation experiments of figure 1. -It is unclear and confusing why the authors compare +Mon & 1x sEV to + 3x sEVs.The authors should also show the + 1x sEVs data.If the method is insensitive in the low range numbers of (spiked) EVs, this should be reported -The CITE-seq analyses should be better described.What are the 'sEV marker-positive droplets'?Are these droplets without cells but with (a certain amount of?)CD9 and/or CD63 protein signal?Deconvolution of SEVtras: -Methods for the experiments shown in figure 2a-c  Minor: The authors have now changed the term 'exosomes' into small EVs (sEVs).Yet, they should explain how they define this subset in their studies.
-The authors define the 2000 x g EVs as large EVs.However, EVs sedimenting at 10,000 x g are generally also defined as large EVs.Why did the authors not analyze this subset?It is very likely a subset that occurs in the sc-RNAseq analyses.
-Lane 83: the used databases are not specifically for small EVs (if these are defined as the EVs that sediment at 100,000 x g).
Reviewer #4 (Remarks to the Author): The revisions have effectively addressed my major concerns regarding the study.I am pleased to see that the imputation has improved the quality of the data, and the investigation of multiple cell lines has demonstrated SEVtras' ability to deconvolute different cell sources.
Although I have only a few minor suggestions: 1.It may be helpful to provide an overall statistic on sEV-containing droplets from the SEVtras analysis for each scRNA-seq dataset examined in this study.Overall, I believe that SEVtras could be an exceptionally powerful tool for extracting biological insights and discoveries from the ever-increasing scRNA-seq studies, and may promote dedicated investigations, including potential biomedical applications.Just curiosity, I am not sure on the deconvoluting capacity of ESAI scores.Since they are numerical values, would it be a problem for deconvoluting complex tissues?Does it make more sense to use a characteristic vector for deconvolution?
Response: Extended Data Fig. 1ab showed a large number of sEVs are present in the tissue.Due to scRNA-seq sample preparation requires to be very fast to ensure cell viability, there was no time for cells to secrete enough sEVs.Hence, existing sEVs will contribute to the majority of sEVs in the scRNA-seq sample.To more clearly track the retention of these previously existing sEVs during scRNA-seq processing, we spiked NanoLuc-labelled sEVs to represent them.We found that a substantial proportion of those sEVs would be retained in the scRNA-seq sample (Fig. 1a and Extended Data Fig. 1ef).
-The methodology used in extended data fig 1 c-e is expression of nano-luc in cells and analyzing nano-luc signals in supernatants.However, this nano-luc can be released by cells in EV-associated and non-EV-associated forms, and is therefore not indicative for EV release.To prevent this, the nano-luc should be coupled to proteins frequently sorted into EVs (eg tetraspanins, as published before by several groups).
Response: Actually, we have conducted sequential differential ultracentrifuge procedures to isolate Nanoluc-labelled sEVs.We have stated more details of Nano-Luc labelled sEVs in the Methods section "NanoLuc-labelled sEVs preparation".
To verify the specificity of NanoLuc protein in sEVs, we first performed NTA analysis.As shown in Extended Data Fig. 1c, the size distribution of NanoLuc-labelled sEVs matched the range of sEVs.To further eliminate the effect of non-sEV-associated forms, we added Triton X and Proteinase K treatments to these sEVs and evaluated NanoLuc luminescence.We found no change in luminescence when NanoLuc-labelled sEVs were treated with Triton X-100 or Proteinase K alone (Extended Data Fig. 1d).The luminescence disappeared only when treated with both Triton X-100 and Proteinase K, suggesting that NanoLuc protein is specifically localized in sEVs.
Performance evaluation of SEVtras: In figure 1b-c the authors included wet lab experiments to validate their technique.Yet, the selected approaches raise several questions.-The monensin treatment protocol should be added to the Methods section Response: Thank you for this suggestion.We have added details in the Methods section "sEV secretion activity stimulation".
-The Methods section should also contain a detailed section on how the sc-RNAseq experiments were performed: how were the cells harvested, how many cells were processed, which washing steps were performed, how were libraries prepared, etc.
Response: Thanks.We have added details in the Methods section "Single cell RNA-seq processing".
-The Methods section should also report to which number of MSC the EVs from 1e4 MSC were added to do the spike-in sc-RNAseq evaluation experiments of figure 1.
Response: Thanks for the suggestion.To ensure consistency, all samples in scRNA-seq were set as 7,000 cells, including the mixture sample of MSC and 293F (3,500 MSCs + 3,500 293F cells).
-It is unclear and confusing why the authors compare +Mon & 1x sEV to + 3x sEVs.The authors should also show the + 1x sEVs data.If the method is insensitive in the low range numbers of (spiked) EVs, this should be reported Response: To assess the impact of ESAI with low range number of sEVs, we compared the sample of "+Mon & 1x sEVs" to "+ Mon alone".SEVtras was able to detect an increase in ESAI between the two samples.In addition, we found that the increase was about 1/3 of the increase in the "+ 3x sEVs" treatment, indicating that SEVtras is reliable in the sEV secretion activity measurement.We have reworded the related part for better understanding.
-The CITE-seq analyses should be better described.What are the 'sEV marker-positive droplets'?Are these droplets without cells but with (a certain amount of?)CD9 and/or CD63 protein signal?

Response:
We have described it more clearly in the revised manuscript.The "sEV marker-positive droplets" are droplets that are positive for either CD9 or CD63.

Deconvolution of SEVtras:
-Methods for the experiments shown in figure 2a-c   Response: Thanks for pointing this out.We have modified our manuscript to clarify that we have set a threshold for the sEV signal score.Response: As suggested, we have added a detailed description.

In
Overall, I believe that SEVtras could be an exceptionally powerful tool for extracting biological insights and discoveries from the ever-increasing scRNA-seq studies, and may promote dedicated investigations, including potential biomedical applications.Just curiosity, I am not sure on the deconvoluting capacity of ESAI scores.Since they are numerical values, would it be a problem for deconvoluting complex tissues?Does it make more sense to use a characteristic vector for deconvolution?
Response: We appreciate the reviewer for these supportive comments.SEVtras has provided customized parameters to achieve this.

Final Decision Letter:
Dear Fangqing, I am pleased to inform you that your Article, "SEVtras delineates small extracellular vesicles at droplet resolution from single-cell transcriptomes", has now been accepted for publication in Nature Methods.Your paper is tentatively scheduled for publication in our * print issue, and will be published online prior to that.The received and accepted dates will be 1st Nov, 2022 and 30th Oct, 2023.This note is intended to let you know what to expect from us over the next month or so, and to let you know where to address any further questions.
Acceptance is conditional on the data in the manuscript not being published elsewhere, or announced in the print or electronic media, until the embargo/publication date.These restrictions are not intended to deter you from presenting your data at academic meetings and conferences, but any enquiries from the media about papers not yet scheduled for publication should be referred to us.
Over the next few weeks, your paper will be copyedited to ensure that it conforms to Nature Methods style.Once your paper is typeset, you will receive an email with a link to choose the appropriate publishing options for your paper and our Author Services team will be in touch regarding any additional information that may be required.
You will receive a link to your electronic proof via email with a request to make any corrections within 48 hours.If, when you receive your proof, you cannot meet this deadline, please inform us at rjsproduction@springernature.com immediately.
Please note that <i>Nature Methods</i> is a Transformative Journal (TJ).Authors may publish their research with us through the traditional subscription access route or make their paper immediately open access through payment of an article-processing charge (APC).Authors will not be required to make a final decision about access to their article until it has been accepted.<a href="https://www.springernature.com/gp/open-research/transformative-journals">Find out more about Transformative Journals</a> Once your manuscript is typeset and you have completed the appropriate grant of rights, you will receive a link to your electronic proof via email with a request to make any corrections within 48 hours.If, when you receive your proof, you cannot meet this deadline, please inform us at rjsproduction@springernature.com immediately.
Once your paper has been scheduled for online publication, the Nature press office will be in touch to confirm the details.
If you have posted a preprint on any preprint server, please ensure that the preprint details are updated with a publication reference, including the DOI and a URL to the published version of the article on the journal website.
Once your paper has been scheduled for online publication, the Nature press office will be in touch to confirm the details.
Content is published online weekly on Mondays and Thursdays, and the embargo is set at 16:00 London time (GMT)/11:00 am US Eastern time (EST) on the day of publication.If you need to know the exact publication date or when the news embargo will be lifted, please contact our press office after you have submitted your proof corrections.Now is the time to inform your Public Relations or Press Office about your paper, as they might be interested in promoting its publication.This will allow them time to prepare an accurate and satisfactory press release.Include your manuscript tracking number NMETH-A50796C and the name of the journal, which they will need when they contact our office.
About one week before your paper is published online, we shall be distributing a press release to news organizations worldwide, which may include details of your work.We are happy for your institution or funding agency to prepare its own press release, but it must mention the embargo date and Nature Methods.Our Press Office will contact you closer to the time of publication, but if you or your Press Office have any inquiries in the meantime, please contact press@nature.com.
To assist our authors in disseminating their research to the broader community, our SharedIt initiative provides you with a unique shareable link that will allow anyone (with or without a subscription) to read the published article.Recipients of the link with a subscription will also be able to download and print the PDF.
As soon as your article is published, you will receive an automated email with your shareable link.
should be described in full experimental detail in the Methods section -Why do the EV plots in figure 2b right panel look different from the ones in figure 2a right panel?What does this say about the methodology or the biology?

2.
In line 79, could you please clarify how droplets containing both cell debris and sEVs are handled?3.In Fig 1b, it is not clear which dots represent the SEVtras analysis and which ones represent experimentally resolved debris.
should be described in full experimental detail in the Methods section Response: We have added experimental details in the Methods section "Single cell RNA-seq processing".-Why do the EV plots in figure 2b right panel look different from the ones in figure 2a right panel?What does this say about the methodology or the biology?Response: Thanks.The difference may relate to cell states because these cells of the sample comes from different plate.After BBKNN batch adjustment with default parameters (Polanski et al. 2019), we found that sEVs from the two kinds of samples showed similar distribution.Please refer to Extended Data Fig. 6b.

2.
In line 79, could you please clarify how droplets containing both cell debris and sEVs are handled?
Fig 1b, it is not clear which dots represent the SEVtras analysis and which ones represent experimentally resolved debris.