Unsupervised gene selection for predicting cell spatial positions in the Drosophila embryo

Yang Chen; Disheng Mao; Yuping Zhang; Zhengqing Ouyang

doi:10.12688/f1000research.20446.2

Home Browse Unsupervised gene selection for predicting cell spatial positions...

ALL Metrics

Views

Downloads

Get PDF

Get XML

Export

▬

✚

Method Article

Revised

Unsupervised gene selection for predicting cell spatial positions in the Drosophila embryo

[version 2; peer review: 1 approved, 1 approved with reservations, 3 not approved]

Previously titled: An unsupervised learning method for reconstructing cell spatial organization with application to the DREAM Single Cell Transcriptomics Challenge

Yang Chen¹, Disheng Mao², Yuping Zhang², Zhengqing Ouyang ¹

PUBLISHED 07 Jun 2021

Author details Author details

¹ Department of Biostatistics and Epidemiology, School of Public Health and Health Sciences, University of Massachusetts, Amherst, MA, USA
² Department of Statistics, University of Connecticut, Storrs, CT, USA

Yang Chen
Roles: Data Curation, Formal Analysis, Investigation, Methodology, Software, Validation, Visualization, Writing – Original Draft Preparation

Disheng Mao
Roles: Investigation, Methodology

Yuping Zhang
Roles: Funding Acquisition, Investigation, Methodology, Resources, Supervision, Writing – Review & Editing

Zhengqing Ouyang
Roles: Conceptualization, Formal Analysis, Funding Acquisition, Investigation, Methodology, Project Administration, Resources, Supervision, Writing – Review & Editing

OPEN PEER REVIEW

REVIEWER STATUS

Abstract

Analyzing single cell RNA-seq data is important for deciphering the spatial relationships, expression patterns, and developmental processes of cells. Combining in situ hybridization-based gene expression atlas images, some works have successfully recovered spatial locations of cells in zebrafish and Drosophila embryos. In this article, we describe a highly ranked method in the DREAM Single Cell Transcriptomics Challenge for predicting cell positions in the Drosophila embryo. The method performs unsupervised feature extraction to select a small number of driver genes and then uses them to predict gene expression and spatial position of each individual cell. First, hierarchical clustering is used to select a subset of driver genes. Second, the similarity matrix of single cells in the bins of the reference atlas is computed. Based on the similarity matrix, the spatial positions of cells are then determined by hierarchical clustering. This method is evaluated on the cell positions and gene expressions in the DREAM Single Cell Transcriptomics Challenge. The comparison with the “silver standard” suggests that our method is effective in reconstructing the cell spatial positions and gene expression patterns in tissues.

Keywords

Spatial organization, single cell RNA-seq, Drosophila embryo, clustering, DREAM challenge

Corresponding author: Zhengqing Ouyang

Competing interests: No competing interests were disclosed.

Grant information: This work was partially supported by the Faculty Research Excellence Program Award at UConn (to YZ).
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Copyright: © 2021 Chen Y et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

How to cite: Chen Y, Mao D, Zhang Y and Ouyang Z. Unsupervised gene selection for predicting cell spatial positions in the Drosophila embryo [version 2; peer review: 1 approved, 1 approved with reservations, 3 not approved]. F1000Research 2021, 9:124 (https://doi.org/10.12688/f1000research.20446.2) First published: 19 Feb 2020, 9:124 (https://doi.org/10.12688/f1000research.20446.1) Latest published: 07 Jun 2021, 9:124 (https://doi.org/10.12688/f1000research.20446.2)

Revised Amendments from Version 1

The title is modified to avoid ambiguity.
The abstract is modified to improve the clarity of method summary.
More citations and description of existing methods are added.
All figures and legends are modified to describe the results more clearly.
More details are added on binarization of scRNA-seq data and prediction of spatial gene expression.
More discussions are added on the interpretation of the results as well as the advantage and limitation of the method.

See the authors' detailed response to the review by Xianwen Ren
See the authors' detailed response to the review by Mark S. Cembrowski and Larissa Kraus

Introduction

The development of single cell RNA sequencing (scRNA-seq) has provided a powerful solution for building a global transcription landscape of all cell types in tissues, finding new cell types, cell lineage tracing, spatial reconstruction, and combining with other omics^1–9. Single cells are originally made from dissociated tissues without spatial information, and spatial gene expression pattern is unknown. In situ hybridization (ISH) and its variants can detect the spatial information of mRNA transcripts and produce gene expression reference atlas. Using enough marker genes, users can reconstruct the spatial position of single cell RNA-seq data by combing the ISH reference atlas^10–15. Some works have also combined sequential fluorescence in situ hybridization (seqFISH) and multiplexed error robust fluorescence in situ hybridization (MERFISH) with scRNA-seq data to map cell types to the reference atlas^16–18.

Recent methods have successfully mapped cells from scRNA-seq data to the spatial positions using dozens of landmark genes^10–15. Nikos et al. developed a DistMap method for mapping the ~1,300 Drosophila embryo cells into ~3,000 bins in the spatial position using 84 marker genes by computing the Matthews Correlation Coefficient (MCC) score between in situ gene expression and binarized scRNA-seq data¹⁰. Satija et al. mapped 851 cells of zebrafish embryo into 64 bins of the spatial embryo with 47 genes. They used a bimodal mixture model to binarize the scRNA-seq data and predict a cell originated from the bin by posterior probability¹¹. Kaia et al. computed the correspondence score and mapped 139 cells into a spatial Platynereis dumerilii brain using a set of 98 genes by whole-mount in situ hybridization (WMISH). They binarized the scRNA-seq data and defined a special score to characterize the relation between every cell-voxel combination. They suggested that 50–100 in situ genes were needed to map cells to spatial position with high confidence using simulated spatial reference atlases¹². Andreas et al. also reconstructed spatial single enterocytes along the villus axis in 1-D space using 102 bottom and top landmark genes from single-molecule fluorescence in situ hybridization (smFISH¹⁹) for 1,383 single cells. They used the summed expression of landmark genes to infer the locations along the 1-D coordinates and tried 50 and 20 landmark genes for prediction. The reconstruction errors were relatively low for 50 genes, but the prediction errors increased remarkably using 20 landmark genes¹³. Nitzan et al. also proposed a new method named novoSpaRc for spatially mapping the scRNA-seq cells into an existing reference atlas. Based on the optimal transport method, it reported the gene expression and spatial position with at most 2 landmark genes for virtual scRNA-seq data and virtual Drosophila embryo¹⁴. The performance of novoSpaRc is still not good for real Drosophila embryo when the marker gene number is smaller than 30^14,15. In the methods mentioned above, the dimension and resolution of the spatial region, as well as the number of marker genes, are key factors that affect the recovery of the spatial position. Although these methods demonstrated good performance, they do not focus on markers genes selection in which biological information were used to determine the spatial gene expression patterns.

To explore new algorithms that include biological information contained in the genes to determine embryonic patterning, the DREAM Single Cell Transcriptomics Challenge asked participating teams to predict the positions in the embryo of 1,297 cells using the expression patterns of 60 (sub-challenge 1), 40 (sub-challenge 2), and 20 (sub-challenge 3) driver genes from the in situ hybridization data²⁰. The challenge endeavors to use less marker genes to infer the spatial locations of cells.

In this article, we assume that cells with similar gene expression will have close spatial distances. By clustering the possible spatial bins for a cell, we select the most possible cluster as the predicted spatial position. The details of our method are the following: First, we use hierarchical clustering for gene selection. Then, we map the cells to spatial positions by computing the mapping similarity between the binarized scRNA-seq and in situ hybridization data and cluster the bins. Finally, we select the optimal thresholds by computing the maximum MCC score for spatial gene expression prediction. We detail our solution for the DREAM challenge, and validate the results using the “silver standard” derived from DistMap. The paper is organized as follows: in Methods, we briefly describe the solutions for all the three sub-challenges; in Results, we present the results of the three sub-challenges in the DREAM Single Cell Transcriptomics Challenge; finally, we discuss and summarize our work.

Methods

Dataset

The scRNA-seq dataset is from ~1,000 handpicked stage 6 fly embryos using Drop-seq²¹. It contains both raw and normalized UMI counts with 1,297 cells and 8,924 genes per cell. In situ hybridization expression patterns of 84 driver genes are from the Berkeley Drosophila Transcription Network Project (BDNTP). The BDNTP reference atlas are binarized. The bin number of one half of the symmetric Drosophila melanogaster embryo at stage 6-pre-gastrulation (which is used in this study) is 3,039. The spatial coordinates of these bins are also specified. The dataset files can be downloaded from the DREAM Single Cell Transcriptomics Challenge after registration with Synapse free of charge (https://www.synapse.org/#!Synapse:syn16782360).

We directly use the normalized scRNA-seq data, the in situ matrix and the geometry of the embryo. The gene names “ E.spl.m5.HLH” and “ Blimp.1” are replaced by “ E(spl)m5-HLH” and “ Blimp-1”.

The DistMap “silver standard” is the best available cell position reference, which is determined by calculating the maximum similarity for cells and spatial bins using all the 84 in situ driver genes in BDNTP. It is shown that the 84 driver gens are sufficient to uniquely and individually label most of the 1,297 cells¹⁰.

Gene selection

We use a hierarchical clustering method to select 60, 40, and 20 driver genes from the 84 genes based on the normalized scRNA-seq data (Figure 1). Specifically, based on the belief that the scRNA-seq gene expression pattern is driven by the driver genes’ expression pattern, we propose to select the essential driver genes based on the information provided by scRNA-seq data. If two genes have high correlation in the scRNA-seq data, we choose only one of them without losing too much of the information. To find the correlated genes, we perform hierarchical clustering on the normalized scRNA-seq data to separate all 84 genes into 60 clusters (with the Euclidean distance and the Mcquitty linkage). The Mcquitty linkage gives more weights for objects in small clusters than those in large clusters in calculating the distance between two clusters. Thus, it is suitable for situations with many small clusters. Since the numbers of clusters are fairly large in the sub-challenge 1 and 2, we opt to use the Mcquitty linkage for distance calculation. In sub-challenge 3, since the total number of clusters is shrunk to 20, which is smaller than sub-challenge 1 and 2, we choose to use the ward linkage in the hierarchical clustering part to obtain larger-sized clusters from the data. After this step, the gene selection process remains the same as sub-challenge 1 and 2.

Figure 1. Workflow of our spatial position prediction method for scRNA-seq with in situ reference gene panel.

(1) Use hierarchical clustering (HC) to select 60/40/20 driver genes from the given 84 driver genes based on scRNA-seq data with the Euclidean distance and the Mcquitty (for 60 and 40 genes) and ward linkage (for 20 genes). (2) Apply the binarization method for the selected genes by minimizing root-mean-square error between of Pearson correlation matrix between the binarized scRNA-seq and the in situ matrix, and use the same criteria to binarize the remaining 24/44/64 predicted genes. (3) Calculate mapping similarity matrix from single cells to spatial bins based on the proportion of matched gene expression between each cell and each bin. (4) For each cell, select top bins based on mapping similarity values by quantile thresholding. Then do HC on the spatial distance of these top bins and select the number of clusters based on the Silhouette score. The top 10 predicted bins are the ones surrounding the largest cluster center. (5) For each gene, we compute the MCC score between predicted gene expression and binarize gene expression given different thresholds. The optimal threshold is corresponding to the maximum MCC. Finally, we apply the optimal threshold for the computeVISH() method in DistMap to predict gene expression. The black points in the scatter plot are the predicted MCC. The red points are Pearson correlation between predicted in situ gene expression with in situ gene expression.

After getting the clusters, we pick the most representative gene of each cluster by calculating the distance between each member gene and the cluster center based on the Euclidean distance and selecting the closest one.

Binarization of scRNA-seq data

We perform binarization on the normalized scRNA-seq data for the selected genes based on the “binarizeSingleCellData()” function in DistMap (https://github.com/rajewsky-lab/distmap). The details of binarization is as follows: for each quantiles from 0.15 to 0.5 with 0.01 as interval, we perform binarization on the scRNA-seq data for 60, 40, and 20 driver genes in sub-challenge 1, 2, 3, respectively. If the gene expression value is larger than the quantile gene expression value, it will be set as 1, otherwise it is 0. Then we compute the difference between the correlation matrix of the binarized scRNA-seq data and that of the in situ matrix based on the root-mean square error. Finally, we select the quantile threshold which has the smallest root mean square error.

Next, to predict the spatial gene expression for all the 84 in situ genes, we used a similar method to binarize the remaining 24, 44, and 64 in situ genes in sub-challenges with the binarized 60, 40, and 20 binarized gene expression. First, we compute the Pearson correlation matrix C₁ between the selected genes (60, 40, and 20) and the remaining genes (24, 44, and 64) using the normalized scRNA-seq data. Second, we binarize the remaining genes based on a series of quantiles from 0.15 to 0.5 with 0.01 as interval. Furthermore, we calculate the Pearson correlation matrix C₂ between the binarized selected genes and the remaining genes. Finally, we select the best quantile for binarizing the remaining genes by minimizing the root mean square error between the Pearson correlation matrices C₁ and C₂.

Compute the similarity matrix between cells and bins

Given the binarized scRNA-seq data and the in situ matrix, we calculate the similarity matrix between the cells and bins based on the selected driver genes. Here, we denote the number of selected driver genes is n_g. The similarity between a cell c_i (i ∈ [1,1,297]) and a bin b_j (j ∈ [1,3,039]) is calculated as follows:

p_{i j} = \frac{n_{s}^{i j}}{n_{g}}, (1)

where $n_{s}^{i j}$ is the number of the same gene expression values (0 or 1) in the two binarized vectors corresponding to cell c_i and bin b_j, and n_g is the in situ matrix for cell c_i and bin b_j.

Select candidate the cell positions based on the similarity matrix

Select top bins. The likelihood of a cell mapped to a bin is determined by the gene expression in the bin and the cell. The bins with the higher similarity are possibly the potential cell position. To generate stable results, we select enough bins (see below) based on the similarity values. Then we use clustering to determine a more refined cell position.

To select the potential bins for predicting cell position, we check the number of bins with high similarity values (> 0.8) in the similarity matrix for each cell from sub-challenge 1, 2, and 3 (Figure 2). The average number of bins with mapping similarity > 0.8 increases as the number of driver genes decreases. In sub-challenge 3, there are more bins with high similarity values produced. The small similarity difference among these bins makes it harder to determine real spatial bins where the cell originated. Thus, as the number of driver genes decreases, the accuracy of bins with high similarity values also decreases. Here, we use the third quartile of the similarity values to select the top bins in sub-challenge 1 with 60 driver genes. We use the first quartile of similarity values in sub-challenge 2 with 40 driver genes. And we use all bins in sub-challenge 3 with 20 driver genes. If the number of selected bins is 0, the bin which has the maximum similarity will be the predicted position. If the number of the selected bins is larger than 100, then only the top 100 bins will be kept based on the similarity values.

Figure 2. Compare the number of high similarity values (> 0.8) in the similarity matrix in each bin for sub-challenge 1, 2, and 3.

The numbers of high similarity values in sub-challenge 1 and 2 are much smaller than those in sub-challenge 3.

Silhouette score to determine the number of clusters of bins

For bins with high similarity values, we need to perform clustering to select the cluster which has the maximum sum of similarity as the predicted cell position. Here, we use hierarchical clustering on the selected bins based on their three-dimensional distance. The cluster number is determined by the silhouette score, which measures the average distance of a point to other points in its cluster compared to the smallest average distance to other clusters. The silhouette score ranges from -1 to +1. The higher the silhouette score, the closer the point is closer to its own cluster and the farther it is away from other clusters. We use the average silhouette score across all selected bins to select the optimal clustering number. Here, we use the NbClust package²² to perform hierarchical clustering with the “centroid” method. Based on the silhouette score, we obtain the number of clusters.

Predict cell positions based on clustering of bins

We compute the sum of similarities for each cluster and select the cluster with the maximum value. Finally, we assign 10 nearest bins to the selected cluster center as the top 10 most possible cell positions.

Predict spatial gene expression based on the similarity matrix

To predict the final spatial in situ gene expression, we use the method computeVISH() in DistMap. It multiplies the normalized gene expression matrix (gene X cell) with the mapping similarity matrix (cell X bin) and sets a threshold to binarize final spatial gene expression. For sub-challenge 1, 2, 3, we select an optimal threshold. First, we generate a series of thresholds between 0 and 1. For each threshold, we compute the predicted spatial gene expression using computeVISH(), then compute the MCC score between the predicted gene expression and the binarized expression of the driver genes. Finally, we select the best threshold based on the largest MCC score to compute the final spatial gene expression.

To estimate the performance of our method to predict spatial gene expression, we compute the Pearson correlation between the in situ gene expression and the predicted gene expression.

Performance evaluation

To evaluate the performance of our method, we use the three performance scores in the DREAM challenge (https://www.synapse.org/#!Synapse:syn17091286, https://github.com/dream-sctc/Scoring/blob/master/dream_scoring_clean.R). The first score (score 1) is the primary score to evaluate the precision of the assignment for the single cells. The second score (score 2) is the average of the relative assignment precision over all the single cells which is used when the first scores are equal for two methods. The third score (score 3) is to compare predictions of gene patterns.

Ambiguous cells: If the predicted top 1 and top 2 positions are the same in the DistMap results, the prediction position will be ambiguous, and the cell will not be computed in the score 1–3. In this challenge, the number of ambiguous cells derived from DistMap are 287.

Results

Select driver genes

We calculated the sums of gene expression values in the in situ matrix for each bin of the 3,039 bins. Each bin has at least one gene expressed in the in situ matrix. It suggests that our selected driver genes cover all bins as at least one driver gene is expressed in each bin. As the number of expressed driver genes decreases when the number of selected driver genes decreases (from 60 to 40 to 20), the prediction accuracy of cell spatial positions also decreases (Figure 3(a)). We also compared the overlaps when varying the number of the driver genes in Figure 3(b). Among the 40 driver genes of the sub-challenge 2, only one driver gene is not in the selected 60 driver genes of the sub-challenge 1. There are 14 overlapped driver genes between the sub-challenge 1, 2 and 3. It suggests our method is consistent on selecting different number of driver genes.

Figure 3.

(a) Compare number of expressed driver genes in each bin in the in situ matrix in sub-challenge 1, 2, and 3 under different selected driver genes scenarios. (b) The Venn diagram of the selected 60, 40, and 20 gene in the three sub-challenges 1, 2, 3. There are 14 common genes in three sub-challenges 1, 2 and 3 (labeled as gene60, gene40, and gene20).

Evaluate the predicted position and spatial gene expression

We used the score 1–3 to evaluate our method for different number of selected driver genes. Figure 4(a) shows the scores of our submitted results for the sub-challenge 1, 2, 3. The blue bar is the score for the silver standard method using 84 driver genes from DistMap. For score 1, the primary score for cell position prediction, the performance of our method is close to the silver standard when using 60 driver genes. The performance decreases when using 40 and 20 driver genes. For score 2, our method shows high scores when using 60 and 40 driver genes and lower score when using 20 driver genes. As Score 2 is the average relative precision for all cells, it suggests that our method is robust for predicting the right position using 60 and 40 driver genes, probably because our algorithm uses the top 10 bins around the largest cluster centers. The score 3 estimates the predicted gene expression patterns. It shows a small difference in our method when using 60 and 40 driver genes. To check the effect of threshold on the prediction results, we test our method under different thresholds. Figure 4(b)–(d) shows the consistency of the score 1–3 over a range of thresholds in the scenarios of different numbers of driver genes. Overall, the results show that our method performs well when using 60 and 40 genes.

Figure 4.

(a) Comparing the score 1, 2, 3 for sub-challenge 1 (60 driver genes), 2 (40 driver genes) and 3 (20 driver genes) with the silver standard (84 driver genes). (b-d) Comparing the score 1, 2, 3 using different thresholds (min, 1st quantile, median, mean, 3rd quantile, max value) for (b) sub-challenge 1 with 60 genes; (c) sub challenge 2 with 40 genes; (d) sub-challenge 3 with 20 genes.

As shown in Figure 5, we used boxplots to present the spatial gene expression prediction accuracy of the 84 driver genes by computing the Pearson correlation between the predicted and in situ gene expression. Consistent with Figure 4(a), the average Pearson correlation of our prediction (60 or 40 driver genes) is higher than that of the silver standard (84 driver genes by DisMap). When the number of driver genes decreases to 20, the average Pearson correlation of our prediction becomes smaller but close to that of the silver standard, which used 84 driver genes. From the results, it suggests that our optimal threshold selection may improve the spatial gene expression prediction. It is feasible to use less genes for spatial gene expression prediction.

Figure 5. Comparing Pearson correlation of predicted spatial gene expression and in situ spatial gene expression for 84 genes between DistMap (84 genes) with our method using (a) 60 genes in subchallenge 1; (b) 40 genes in sub-challenge 2; and (c) 20 genes in sub-challenge 3.

For each gene, we calculate the Pearson correlation between the predicted spatial gene expression and the in situ spatial gene expression. The average Pearson correlation value of DistMap (84 genes), sub-challenge 1, sub-challenge 2, and sub-challenge 3 are 0.532, 0.622, 0.561, and 0.422, respectively.

Discussion

With high quality Drosophila dataset, Karaiskos et al. mapped single cells to the 3D embryo positions using the DistMap method¹⁰. It assumed that transcriptomes of neighboring cells should show similar patterns in spatial tissues. We further assumed that cells with similar gene expression will have close spatial distance. We applied hierarchical clustering to select a small number of driver genes and group similar cells to be mapped to the 3D embryo. After selecting the largest cluster, we determined the most possible predicted spatial cell position around the cluster center.

We assessed the performance of our method on the selected 60, 40 or 20 driver genes using score 1–3 and compared it with the silver standard (DistMap results). For 60 driver genes, our results show close performance to the silver standard in score 1 and 3, and higher performance in score 2. It suggests that our method applied to 60 driver genes can achieve the same or higher level of performance as the silver standard to predict cell position. Our method obtains better performance for gene expression pattern prediction with 60 and 40 genes. When using 40 driver genes, the score 1 of our method decreases a little but the score 3 is very close to the silver standard. It suggests that our method can still predict gene expression pattern reasonably well using 40 genes. Fore 20 driver genes, the performance of our method is lower than but still close to the silver standard in both cell position and gene expression prediction.

Overall, our method achieves good performance to predict spatial cell position and gene expression with 60 and 40 driver genes. The primary reason may be that our method selects the most representing genes in the gene clusters for prediction. Also, for potential spatial bins, our method utilizes bin clusters and selects the cluster center as predicted position. The good performance of our method is consistent to our assumption that cells with close spatial position show similar gene expression patterns. The performance of our method decreases when the number of driver genes reduces to 20. This is probably because the similarities between cells and spatial bins have larger variations when measured using smaller number of genes. In this case, it becomes harder to find the most possible bin position based on cell-bin similarities. In Nitzan’s work^14,15, they successfully reconstructed the virtual Drosophila embryo with one marker gene and the structural information between gene expression space and physical space. It suggested that gene-gene correlation and spatial correlation contain useful information for gene expression and spatial cell position prediction. In addition, our method can be extended using an iterative approach by updating the mapping similarity matrix at each step.

Data availability

The dataset associated with the DREAM Single Cell Transcriptomics Challenge is available at https://www.synapse.org/#!Synapse:syn16782360 and https://www.synapse.org/#!Synapse:syn18632189. Limited by the sharing settings of Synapse, users should register in Synapse (free of charge; https://www.synapse.org/) using their email address, and agree to the dataset conditions of use (https://www.synapse.org/#!Synapse:syn15665609/wiki/583240).

Software availability

Source code implementation for the method presented in this article and used in the DREAM Single Cell Transcriptomics Challenge is available at: https://github.com/ouyang-lab/SCTC-Challenge-zho_team. The scoring scripts are available at: https://github.com/dream-sctc/Scoring/blob/master/dream_scoring_clean.R.

Archived source code as at time of publication can be accessed at: https://doi.org/10.5281/zenodo.3592532

License: GLP 3.0

Author information

All authors actively contributed to the results and discussed the procedures. They have been involved in the test, design and evaluation of various approaches; many of the explored techniques were not used in the final workflow due to strong overfitting. DM: gene selection and coding. YC: data pre-processing, clustering, and writeup. ZO: supervision, realization of workflows, and writeup. YZ: supervision, realization of workflows, and writeup.

Acknowledgements

Data used in this publication were generated by Prof. Dr. Nikolaus Rajewsky, Max Delbrück at the Center for Molecular Medicine, and these results were obtained as part of the DREAM Single Cell Transcriptomics Challenge project through Synapse ID (syn15665609). We thank Pablo Meyer and Jovan Tanevski for providing the documents, code and suggestions for score 1–3.

Faculty Opinions recommended

References

1. Farrell JA, Wang Y, Riesenfeld SJ, et al.: Single-cell reconstruction of developmental trajectories during zebrafish embryogenesis. Science. 2018; 360(6392): eaar3131. PubMed Abstract | Publisher Full Text | Free Full Text
2. Butler A, Hoffman P, Smibert P, et al.: Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat Biotechnol. 2018; 36(5): 411–420. PubMed Abstract | Publisher Full Text | Free Full Text
3. Colomé-Tatché M, Theis FJ: Statistical single cell multi-omics integration. Curr Opin Syst Biol. 2018; 7: 54–59. Publisher Full Text
4. Gu C, Liu S, Wu Q, et al.: Integrative single-cell analysis of transcriptome, DNA methylome and chromatin accessibility in mouse oocytes. Cell Res. 2019; 29(2): 110–123. PubMed Abstract | Publisher Full Text | Free Full Text
5. Bian S, Hou Y, Zhou X, et al.: Single-cell multiomics sequencing and analyses of human colorectal cancer. Science. 2018; 362(6418): 1060–1063. PubMed Abstract | Publisher Full Text
6. Kulkarni A, Anderson AG, Merullo Dp, et al.: Beyond bulk: a review of single cell transcriptomics methodologies and applications. Curr Opin Biotechnol. 2019; 58: 129–136. PubMed Abstract | Publisher Full Text | Free Full Text
7. Savulescu AF, Jacobs C, Negishi Y, et al.: Pinpointing cell identity in time and space. Front Mol Biosci. 2020; 7: 209. PubMed Abstract | Publisher Full Text | Free Full Text
8. Stuart T, Satija R: Integrative single-cell analysis. Nat Rev Genet. 2019; 20(5): 257–272. PubMed Abstract | Publisher Full Text
9. Mayr U, Serra D, Liberali P: Exploring single cells in space and time during tissue development, homeostasis and regeneration. Development. 2019; 146(12): dev176727. PubMed Abstract | Publisher Full Text
10. Karaiskos N, Wahle P, Alles J, et al.: The Drosophila embryo at single-cell transcriptome resolution. Science. 2017; 358(6360): 194–199. PubMed Abstract | Publisher Full Text
11. Satija R, Farrell JA, Gennert D, et al.: Spatial reconstruction of single-cell gene expression data. Nat Biotechnol. 2015; 33(5): 495–502. PubMed Abstract | Publisher Full Text | Free Full Text
12. Achim K, Pettit JB, Saraiva LR, et al.: High-throughput spatial mapping of single-cell RNA-seq data to tissue of origin. Nat Biotechnol. 2015; 33(5): 503–9. PubMed Abstract | Publisher Full Text
13. Moor AE, Harnik Y, Ben-Moshe S, et al.: Spatial reconstruction of single enterocytes uncovers broad zonation along the intestinal villus axis. Cell. 2018; 175(4): 1156–1167.e15. PubMed Abstract | Publisher Full Text
14. Nitzan M, Karaiskos N, Friedman N, et al.: Charting a tissue from single-cell transcriptomes. bioRxiv. 2018; 456350. Publisher Full Text
15. Nitzan M, Karaiskos N, Friedman N, et al.: Gene expression cartography. Nature. 2019; 576(7785): 132–137. PubMed Abstract | Publisher Full Text
16. Shah S, Lubeck E, Zhou W, et al.: In situ transcription profiling of single cells reveals spatial organization of cells in the mouse hippocampus. Neuron. 2016; 92(2): 342–357. PubMed Abstract | Publisher Full Text | Free Full Text
17. Zhu Q, Shah S, Dries R, et al.: Identification of spatially associated subpopulations by combining scRNAseq and sequential fluorescence in situ hybridization data. Nat Biotechnol. 2018; 36: 1183–1190. PubMed Abstract | Publisher Full Text | Free Full Text
18. Moffitt JR, Bambah-Mukku D, Eichhorn SW, et al.: Molecular, spatial, and functional single-cell profiling of the hypothalamic preoptic region. Science. 2018; 362(6416): eaau5324. PubMed Abstract | Publisher Full Text | Free Full Text
19. Halpern KB, Shenhav R, Matcovitch-Natan O, et al.: Single-cell spatial reconstruction reveals global division of labour in the mammalian liver. Nature. 2017; 542(7641): 352–356. PubMed Abstract | Publisher Full Text | Free Full Text
20. Tanevski J, Nguyen T, Truong B, et al.: Gene selection for optimal prediction of cell position in tissues from single-cell transcriptomics data. Life Sci Alliance. 2020; 3(11): e202000867. PubMed Abstract | Publisher Full Text | Free Full Text
21. Macosko EZ, Basu A, Satija R, et al.: Highly Parallel Genome-wide Expression Profiling of Individual Cells Using Nanoliter Droplets. Cell. 2015; 161(5): 1202–1214. PubMed Abstract | Publisher Full Text | Free Full Text
22. Charrad M, Ghazzali N, Boiteau V, et al.: NbClust: An R Package for Determining the Relevant Number of Clusters in a Data Set. J Stat Softw. 2014; 61(6): 1–36. Publisher Full Text

Comments on this article Comments (0)

Version 2

VERSION 2 PUBLISHED 19 Feb 2020

Author details Author details

Yang Chen
Roles: Data Curation, Formal Analysis, Investigation, Methodology, Software, Validation, Visualization, Writing – Original Draft Preparation

Disheng Mao
Roles: Investigation, Methodology

Yuping Zhang
Roles: Funding Acquisition, Investigation, Methodology, Resources, Supervision, Writing – Review & Editing

Zhengqing Ouyang
Roles: Conceptualization, Formal Analysis, Funding Acquisition, Investigation, Methodology, Project Administration, Resources, Supervision, Writing – Review & Editing

Competing interests

No competing interests were disclosed.

Grant information

This work was partially supported by the Faculty Research Excellence Program Award at UConn (to YZ).
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Article Versions (2)

version 2

Revised

Published: 07 Jun 2021, 9:124

https://doi.org/10.12688/f1000research.20446.2

version 1

Published: 19 Feb 2020, 9:124

https://doi.org/10.12688/f1000research.20446.1

© 2021 Chen Y et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Download

Export To

metrics

	Views	Downloads
F1000Research	-	-
PubMed Central Data from PMC are received and updated monthly.	-	-

Citations

SEE MORE DETAILS

CITE

how to cite this article

Chen Y, Mao D, Zhang Y and Ouyang Z. Unsupervised gene selection for predicting cell spatial positions in the Drosophila embryo [version 2; peer review: 1 approved, 1 approved with reservations, 3 not approved]. F1000Research 2021, 9:124 (https://doi.org/10.12688/f1000research.20446.2)

NOTE: If applicable, it is important to ensure the information in square brackets after the title is included in all citations of this article.

track

receive updates on this article

Track an article to receive email alerts on any updates to this article.

Open Peer Review

Current Reviewer Status: ?

Key to Reviewer Statuses VIEW HIDE

ApprovedThe paper is scientifically sound in its current form and only minor, if any, improvements are suggested

Approved with reservations A number of small changes, sometimes more significant revisions are required to address specific details and improve the papers academic merit.

Not approvedFundamental flaws in the paper seriously undermine the findings and conclusions

Version 2

VERSION 2

PUBLISHED 07 Jun 2021

Revised

Views

Reviewer Report 28 Jun 2024

Komal Kumar Bollepogu Raja, Baylor college of medicine, Houston, TX, USA

Approved with Reservations

https://doi.org/10.5256/f1000research.53961.r290214

In this manuscript Chen et al develop an unsupervised gene selection model that predicts cell positions. They have utilized published single cell RNA seq dataset from Drosophila embryo. Further, hierarchical clustering was used to select a set of driver genes to predict cell positions. The authors compare their model with Distmap and show that their model performs well with 60 genes. The manuscript is well written and the data is presented well. Spatial reconstruction of tissues using single cell and other omics approaches is an emerging field which allows identifying dynamic gene expression programs across tissues. Therefore, the current study can be a valuable resource. However, the authors have evaluated their model on only a single dataset. It will be worthwhile to evaluate this method on at least one more independent dataset to see its consistency. As of now, the method cannot be generalized to a wide range of tissues.

Minor comment:
1. The last line in the Abstract should be changed: "The comparison with the Ã¢ÂÂsilver standardÃ¢ÂÂ suggests that our method is effective in reconstructing the cell spatial positions and gene expression patterns in tissues." Instead of tissues I suggest Drosophila embryo.

Is the rationale for developing the new method (or application) clearly explained?

Yes
Is the description of the method technically sound?

Yes
Are sufficient details provided to allow replication of the method development and its use by others?

Yes
If any results are presented, are all the source data underlying the results available to ensure full reproducibility?

Yes
Are the conclusions about the method and its performance adequately supported by the findings presented in the article?

Partly

Competing Interests: No competing interests were disclosed.

Reviewer Expertise: Single cell omics, Genomics and Bioinformatics

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

CITE

Report a concern

Respond or Comment

Views

Reviewer Report 28 Jun 2024

Saumitra Dey Choudhury, All India Institute of Medical Sciences, New Delhi, India

Not Approved

https://doi.org/10.5256/f1000research.53961.r290210

In this manuscript titled Unsupervised gene selection for predicting cell spatial positions in the Drosophila embryo, the authors have attempted to computationally deduce the spatial positions of individual cells in Drosophila embryos from their Drop-seq data in reference to the in situ hybridization expression patterns of 84 driver genes from the BDNTP. Though the method is well defined and could be useful in general to map spatial gene expression using transcriptomic data, some concerns exist:

First, the Methods section does not have any explanation of the wet lab procedure of homogenization of embryos, making the single-cell suspension and cDNA library preparation. Second, the quality of the Drop-seq data cannot be established from the manuscript. Third, the word Unsupervised in the title is misleading. Fourth and most importantly, experimental validation of the performance of the method is required; at least for a few of the driver genes using antibodies/ immunofluorescence.

In general, this manuscript brings forth a strong computational method to extrapolate spatial cell positions but lacks the experimental validation to support it.

Is the rationale for developing the new method (or application) clearly explained?

Yes
Is the description of the method technically sound?

Yes
Are sufficient details provided to allow replication of the method development and its use by others?

Yes
If any results are presented, are all the source data underlying the results available to ensure full reproducibility?

No source data required
Are the conclusions about the method and its performance adequately supported by the findings presented in the article?

No

Competing Interests: No competing interests were disclosed.

Reviewer Expertise: Drosophila Neurogenetics, Single-cell RNA Transcriptomics, Confocal and Super-resolution Microscopy

I confirm that I have read this submission and believe that I have an appropriate level of expertise to state that I do not consider it to be of an acceptable scientific standard, for reasons outlined above.

CITE

Report a concern

Respond or Comment

Views

Reviewer Report 27 Jun 2024

Yen-Chung Chen, New York University, New York, New York, USA

Not Approved

https://doi.org/10.5256/f1000research.53961.r290211

Chen et al. described a novel method for feature selection for spatially mapping single-cell transcriptomic data. Originally a leading solution in DREAM Single-Cell Transcriptomics challenge [1], this method has demonstrated its relative edge in the challenge and superior performance in certain aspects against distMap [2]. While being indeed an advance in methodology, the work left several aspect to be desired to generalize and address feature selection in spatial mapping.

First, the original 84-gene pool was pre-selected based on prior knowledge of importance and spatial expression pattern in developing fly embryos [3] while most single-cell transcriptomic atlas are and will be generated from samples lacking prior knowledge with such breadth and resolution. It therefore remains open whether this proposed method will remain efficient when feature selection has to be performed on all genes or all variable genes. To address this issue, the method should be tested on more recent spatial and classic single-cell transcriptomic dataset to benchmark whether the method performs favorably when selecting features from single-cell RNA-seq without supervision from prior knowledge.

Second, since assigning a cell to a bin can be considered as transferring the label of the bin to the cell of interest, potential users would be curious how the method holds up against the broad selection of existing label transfer methods, including Seurat [4], scVI [5], and classic classifiers like multimodal logistic regression, random forest classifier, and support vector machine for predicting which in situ bin a cell will match. It might be of interest that a previous iteration of Seurat used to demonstrate a similar spatial mapping task [6] and will be a great benchmarking dataset.

Finally, there are minor methodological decisions that were reported where discussion of rationale could help a potential user and future method development. Specifically, in section Gene selection, the choice between Ward and Mcquitty linkage appeared arbitrary to me and made me wonder that for a user working on a different dataset, how the decision should be made (e.g., with regard to a fixed total number of clusters?). In section Compute the similarity matrix between cells and bins, it is unclear why p_ij was chosen as the scoring scheme instead of correlation coefficient or mutual information.

In sum, the work represents a novel and efficient method addressing the pressing need to systematically supplement existing single-cell transcriptomic atlases with spatial information, and the authors have demonstrated its power in answering a specific question. However, its general applicability remains to be tested on available spatial transcriptomic datasets collected in other contexts and tissues (e.g., [7-9]).

Is the rationale for developing the new method (or application) clearly explained?

Yes
Is the description of the method technically sound?

Partly
Are sufficient details provided to allow replication of the method development and its use by others?

Yes
If any results are presented, are all the source data underlying the results available to ensure full reproducibility?

No source data required
Are the conclusions about the method and its performance adequately supported by the findings presented in the article?

Partly

References

1. Tanevski J, Nguyen T, Truong B, Karaiskos N, et al.: Gene selection for optimal prediction of cell position in tissues from single-cell transcriptomics data.Life Sci Alliance. 2020; 3 (11). PubMed Abstract | Publisher Full Text
2. Karaiskos N, Wahle P, Alles J, Boltengagen A, et al.: The Drosophila embryo at single-cell transcriptome resolution.Science. 2017; 358 (6360): 194-199 PubMed Abstract | Publisher Full Text
3. Fowlkes CC, Hendriks CL, Keränen SV, Weber GH, et al.: A quantitative spatiotemporal atlas of gene expression in the Drosophila blastoderm.Cell. 2008; 133 (2): 364-74 PubMed Abstract | Publisher Full Text
4. Hao Y, Stuart T, Kowalski MH, Choudhary S, et al.: Dictionary learning for integrative, multimodal and scalable single-cell analysis.Nat Biotechnol. 2024; 42 (2): 293-304 PubMed Abstract | Publisher Full Text
5. Gayoso A, Lopez R, Xing G, Boyeau P, et al.: A Python library for probabilistic analysis of single-cell omics data. Nature Biotechnology. 2022; 40 (2): 163-166 Publisher Full Text
6. Mayer C, Hafemeister C, Bandler RC, Machold R, et al.: Developmental diversification of cortical inhibitory interneurons.Nature. 2018; 555 (7697): 457-462 PubMed Abstract | Publisher Full Text
7. Maniatis S, Äijö T, Vickovic S, Braine C, et al.: Spatiotemporal dynamics of molecular pathology in amyotrophic lateral sclerosis.Science. 2019; 364 (6435): 89-93 PubMed Abstract | Publisher Full Text
8. Langlieb J, Sachdev NS, Balderrama KS, Nadaf NM, et al.: The molecular cytoarchitecture of the adult mouse brain.Nature. 2023; 624 (7991): 333-342 PubMed Abstract | Publisher Full Text
9. Wang M, Hu Q, Lv T, Wang Y, et al.: High-resolution 3D spatiotemporal transcriptomic maps of developing Drosophila embryos and larvae.Dev Cell. 2022; 57 (10): 1271-1283.e4 PubMed Abstract | Publisher Full Text

Competing Interests: No competing interests were disclosed.

Reviewer Expertise: Genomics, developmental biology, biostatistics

CITE

Report a concern

Respond or Comment

Views

Reviewer Report 17 Jun 2021

Xianwen Ren, Biomedical Pioneering Innovation Center, Peking University, Beijing, China

Approved

https://doi.org/10.5256/f1000research.53961.r87000

The authors have ... Continue reading

CITE

Report a concern

Respond or Comment

Version 1

VERSION 1

PUBLISHED 19 Feb 2020

Views

Reviewer Report 23 Sep 2020

Mark S. Cembrowski, Dept. of Cellular and Physiological Sciences, University of British Columbia, Vancouver, BC, Canada

Larissa Kraus, Department of Cellular and Physiological Sciences, University of British Columbia, Vancouver, BC, Canada

Not Approved

https://doi.org/10.5256/f1000research.22477.r70705

In this manuscript, the authors describe an approach to infer spatial gene expression using scRNA-seq data. The authors aimed to identify spatial expression of cell clusters using 20, 40 or 60 marker genes, and compared to the “gold standard” Distmap method using 84 marker genes.

The authors begin to explain the rationale of the manuscript by describing the concept of the DREAM challenge, which aims to introduce novel computational techniques to map spatial gene expression using transcriptomic data. Since the DREAM challenge aims to implement novel methods, it is imperative to introduce and carefully compare common methods with the novel proposed approach, which was underdeveloped by the authors. As an example, the “gold standard” method is mentioned several times, though it is not clear what exactly the “gold standard” is or how well-controlled it is. As another key example, a discussion of any kind is completely missing.

The design of figures for a manuscript should be chosen to clearly emphasize the main points of the data. Here, the figures are difficult to interpret because key information and description is missing (e.g., legends are too brief, axis labels are missing, subplot headings are redundant, figure design is inconsistent). In addition, the design and presentation of the figures seem unmotivated and thereby difficult for the reader to see the importance of any particular data or easily draw conclusions from the results. In addition, consistent colour schemes and designs would help the reader to follow the manuscript and understand the results.

Is the rationale for developing the new method (or application) clearly explained?

Yes
Is the description of the method technically sound?

Yes
Are sufficient details provided to allow replication of the method development and its use by others?

Partly
If any results are presented, are all the source data underlying the results available to ensure full reproducibility?

Yes
Are the conclusions about the method and its performance adequately supported by the findings presented in the article?

No

Competing Interests: No competing interests were disclosed.

Reviewer Expertise: Transcriptomics

We confirm that we have read this submission and believe that we have an appropriate level of expertise to state that we do not consider it to be of an acceptable scientific standard, for reasons outlined above.

CITE

Report a concern

Author Response 07 Jun 2021

Zhengqing Ouyang, Department of Biostatistics and Epidemiology, School of Public Health and Health Sciences, University of Massachusetts, Amherst, USA

07 Jun 2021

Author Response

Point-by-point response to the comments of Reviewer 2

Reviewer 2

In this manuscript, the authors describe an approach to infer spatial gene expression using scRNA-seq data. The authors ... Continue reading Point-by-point response to the comments of Reviewer 2

Reviewer 2

In this manuscript, the authors describe an approach to infer spatial gene expression using scRNA-seq data. The authors aimed to identify spatial expression of cell clusters using 20, 40 or 60 marker genes, and compared to the “gold standard” Distmap method using 84 marker genes.

The authors begin to explain the rationale of the manuscript by describing the concept of the DREAM challenge, which aims to introduce novel computational techniques to map spatial gene expression using transcriptomic data.
Since the DREAM challenge aims to implement novel methods, it is imperative to introduce and carefully compare common methods with the novel proposed approach, which was underdeveloped by the authors. As an example, the “gold standard” method is mentioned several times, though it is not clear what exactly the “gold standard” is or how well-controlled it is. As another key example, a discussion of any kind is completely missing.

We thank the reviewer for the constructive comments. The “gold standard” is generated by DistMap, which is the best available single cell position reference used in the DREAM Challenge. DistMap determines single cell position by searching the maximum similarity for cells and spatial bins using the in situ expression of 84 driver genes in the Berkeley Drosophila Transcription Network Project (BDNTP). It is shown that the 84 driver gens are sufficient to uniquely and individually label most of the 1,297 cells (Karaiskos, et al. 2017). Thus, it is used in the DREAM Challenge to assess the performance of methods using smaller numbers (60, 40, and 20) of driver genes. In the revised manuscript, we changed “gold standard” to “silver standard”, as the cell positions generated by DistMap are not experimentally validated. We added the description of the DistMap “silver standard” in the Method section.

We also added more introductions to the common methods for the cell position prediction from single cell transcriptome data in the Introduction section. We described the limitation of current methods and explain the motivation to study how the driver gene selection affecting the prediction of cell spatial position.

The comparisons of our methods (Zho_team) and other top performing methods have been described in the DREAM Single Cell Transcriptomics Challenge consortium paper (Gene selection for optimal prediction of cell position in tissues from single-cell transcriptomics data, Life Sci Alliance. 2020 Nov; 3(11): e202000867). Our manuscript is focused on describing the details of our method used in the DREAM Challenge.

We also expanded the Discussion section of our manuscript. In addition to summarizing our method and results, we also pointed that gene-gene correlation and spatial correlation are useful information for spatial cell position and gene expression prediction.

The design of figures for a manuscript should be chosen to clearly emphasize the main points of the data. Here, the figures are difficult to interpret because key information and description is missing (e.g., legends are too brief, axis labels are missing, subplot headings are redundant, figure design is inconsistent). In addition, the design and presentation of the figures seem unmotivated and thereby difficult for the reader to see the importance of any particular data or easily draw conclusions from the results.

In addition, consistent colour schemes and designs would help the reader to follow the manuscript and understand the results.

We thank the reviewer for the constructive comments. In the revised manuscript, we improved all figures and legends. We enlarged the font sizes for all figures. In Figure 1, we clarified the whole workflow to display the motivation of our method. In Figure 2, we changed it to display the number of bins with high similarity values (> 0.8) in the similarity matrix for each cell when the number of driver genes is 60, 40, or 20. In Figure 3, we enlarged the font size and added more information in the figure labels. In Figure 4, we enlarged the font size. In figure 5, we changed it to the comparison of the Pearson correlation of predicted gene expression from DistMap (using 84 driver genes) and our method (using 60, 40, or 20 genes).
Point-by-point response to the comments of Reviewer 2

Reviewer 2

In this manuscript, the authors describe an approach to infer spatial gene expression using scRNA-seq data. The authors aimed to identify spatial expression of cell clusters using 20, 40 or 60 marker genes, and compared to the “gold standard” Distmap method using 84 marker genes.

The authors begin to explain the rationale of the manuscript by describing the concept of the DREAM challenge, which aims to introduce novel computational techniques to map spatial gene expression using transcriptomic data.
Since the DREAM challenge aims to implement novel methods, it is imperative to introduce and carefully compare common methods with the novel proposed approach, which was underdeveloped by the authors. As an example, the “gold standard” method is mentioned several times, though it is not clear what exactly the “gold standard” is or how well-controlled it is. As another key example, a discussion of any kind is completely missing.

We thank the reviewer for the constructive comments. The “gold standard” is generated by DistMap, which is the best available single cell position reference used in the DREAM Challenge. DistMap determines single cell position by searching the maximum similarity for cells and spatial bins using the in situ expression of 84 driver genes in the Berkeley Drosophila Transcription Network Project (BDNTP). It is shown that the 84 driver gens are sufficient to uniquely and individually label most of the 1,297 cells (Karaiskos, et al. 2017). Thus, it is used in the DREAM Challenge to assess the performance of methods using smaller numbers (60, 40, and 20) of driver genes. In the revised manuscript, we changed “gold standard” to “silver standard”, as the cell positions generated by DistMap are not experimentally validated. We added the description of the DistMap “silver standard” in the Method section.

We also added more introductions to the common methods for the cell position prediction from single cell transcriptome data in the Introduction section. We described the limitation of current methods and explain the motivation to study how the driver gene selection affecting the prediction of cell spatial position.

The comparisons of our methods (Zho_team) and other top performing methods have been described in the DREAM Single Cell Transcriptomics Challenge consortium paper (Gene selection for optimal prediction of cell position in tissues from single-cell transcriptomics data, Life Sci Alliance. 2020 Nov; 3(11): e202000867). Our manuscript is focused on describing the details of our method used in the DREAM Challenge.

We also expanded the Discussion section of our manuscript. In addition to summarizing our method and results, we also pointed that gene-gene correlation and spatial correlation are useful information for spatial cell position and gene expression prediction.

The design of figures for a manuscript should be chosen to clearly emphasize the main points of the data. Here, the figures are difficult to interpret because key information and description is missing (e.g., legends are too brief, axis labels are missing, subplot headings are redundant, figure design is inconsistent). In addition, the design and presentation of the figures seem unmotivated and thereby difficult for the reader to see the importance of any particular data or easily draw conclusions from the results.

In addition, consistent colour schemes and designs would help the reader to follow the manuscript and understand the results.

We thank the reviewer for the constructive comments. In the revised manuscript, we improved all figures and legends. We enlarged the font sizes for all figures. In Figure 1, we clarified the whole workflow to display the motivation of our method. In Figure 2, we changed it to display the number of bins with high similarity values (> 0.8) in the similarity matrix for each cell when the number of driver genes is 60, 40, or 20. In Figure 3, we enlarged the font size and added more information in the figure labels. In Figure 4, we enlarged the font size. In figure 5, we changed it to the comparison of the Pearson correlation of predicted gene expression from DistMap (using 84 driver genes) and our method (using 60, 40, or 20 genes).
Competing Interests: No competing interests were disclosed. Close
Report a concern
Respond or Comment

COMMENTS ON THIS REPORT

Author Response 07 Jun 2021

Zhengqing Ouyang, Department of Biostatistics and Epidemiology, School of Public Health and Health Sciences, University of Massachusetts, Amherst, USA

07 Jun 2021

Author Response

Point-by-point response to the comments of Reviewer 2

Reviewer 2

In this manuscript, the authors describe an approach to infer spatial gene expression using scRNA-seq data. The authors ... Continue reading Point-by-point response to the comments of Reviewer 2

Reviewer 2

In this manuscript, the authors describe an approach to infer spatial gene expression using scRNA-seq data. The authors aimed to identify spatial expression of cell clusters using 20, 40 or 60 marker genes, and compared to the “gold standard” Distmap method using 84 marker genes.

The authors begin to explain the rationale of the manuscript by describing the concept of the DREAM challenge, which aims to introduce novel computational techniques to map spatial gene expression using transcriptomic data.
Since the DREAM challenge aims to implement novel methods, it is imperative to introduce and carefully compare common methods with the novel proposed approach, which was underdeveloped by the authors. As an example, the “gold standard” method is mentioned several times, though it is not clear what exactly the “gold standard” is or how well-controlled it is. As another key example, a discussion of any kind is completely missing.

We thank the reviewer for the constructive comments. The “gold standard” is generated by DistMap, which is the best available single cell position reference used in the DREAM Challenge. DistMap determines single cell position by searching the maximum similarity for cells and spatial bins using the in situ expression of 84 driver genes in the Berkeley Drosophila Transcription Network Project (BDNTP). It is shown that the 84 driver gens are sufficient to uniquely and individually label most of the 1,297 cells (Karaiskos, et al. 2017). Thus, it is used in the DREAM Challenge to assess the performance of methods using smaller numbers (60, 40, and 20) of driver genes. In the revised manuscript, we changed “gold standard” to “silver standard”, as the cell positions generated by DistMap are not experimentally validated. We added the description of the DistMap “silver standard” in the Method section.

We also added more introductions to the common methods for the cell position prediction from single cell transcriptome data in the Introduction section. We described the limitation of current methods and explain the motivation to study how the driver gene selection affecting the prediction of cell spatial position.

The comparisons of our methods (Zho_team) and other top performing methods have been described in the DREAM Single Cell Transcriptomics Challenge consortium paper (Gene selection for optimal prediction of cell position in tissues from single-cell transcriptomics data, Life Sci Alliance. 2020 Nov; 3(11): e202000867). Our manuscript is focused on describing the details of our method used in the DREAM Challenge.

We also expanded the Discussion section of our manuscript. In addition to summarizing our method and results, we also pointed that gene-gene correlation and spatial correlation are useful information for spatial cell position and gene expression prediction.

The design of figures for a manuscript should be chosen to clearly emphasize the main points of the data. Here, the figures are difficult to interpret because key information and description is missing (e.g., legends are too brief, axis labels are missing, subplot headings are redundant, figure design is inconsistent). In addition, the design and presentation of the figures seem unmotivated and thereby difficult for the reader to see the importance of any particular data or easily draw conclusions from the results.

In addition, consistent colour schemes and designs would help the reader to follow the manuscript and understand the results.

We thank the reviewer for the constructive comments. In the revised manuscript, we improved all figures and legends. We enlarged the font sizes for all figures. In Figure 1, we clarified the whole workflow to display the motivation of our method. In Figure 2, we changed it to display the number of bins with high similarity values (> 0.8) in the similarity matrix for each cell when the number of driver genes is 60, 40, or 20. In Figure 3, we enlarged the font size and added more information in the figure labels. In Figure 4, we enlarged the font size. In figure 5, we changed it to the comparison of the Pearson correlation of predicted gene expression from DistMap (using 84 driver genes) and our method (using 60, 40, or 20 genes).
Point-by-point response to the comments of Reviewer 2

Reviewer 2

In this manuscript, the authors describe an approach to infer spatial gene expression using scRNA-seq data. The authors aimed to identify spatial expression of cell clusters using 20, 40 or 60 marker genes, and compared to the “gold standard” Distmap method using 84 marker genes.

The authors begin to explain the rationale of the manuscript by describing the concept of the DREAM challenge, which aims to introduce novel computational techniques to map spatial gene expression using transcriptomic data.
Since the DREAM challenge aims to implement novel methods, it is imperative to introduce and carefully compare common methods with the novel proposed approach, which was underdeveloped by the authors. As an example, the “gold standard” method is mentioned several times, though it is not clear what exactly the “gold standard” is or how well-controlled it is. As another key example, a discussion of any kind is completely missing.

We thank the reviewer for the constructive comments. The “gold standard” is generated by DistMap, which is the best available single cell position reference used in the DREAM Challenge. DistMap determines single cell position by searching the maximum similarity for cells and spatial bins using the in situ expression of 84 driver genes in the Berkeley Drosophila Transcription Network Project (BDNTP). It is shown that the 84 driver gens are sufficient to uniquely and individually label most of the 1,297 cells (Karaiskos, et al. 2017). Thus, it is used in the DREAM Challenge to assess the performance of methods using smaller numbers (60, 40, and 20) of driver genes. In the revised manuscript, we changed “gold standard” to “silver standard”, as the cell positions generated by DistMap are not experimentally validated. We added the description of the DistMap “silver standard” in the Method section.

We also added more introductions to the common methods for the cell position prediction from single cell transcriptome data in the Introduction section. We described the limitation of current methods and explain the motivation to study how the driver gene selection affecting the prediction of cell spatial position.

The comparisons of our methods (Zho_team) and other top performing methods have been described in the DREAM Single Cell Transcriptomics Challenge consortium paper (Gene selection for optimal prediction of cell position in tissues from single-cell transcriptomics data, Life Sci Alliance. 2020 Nov; 3(11): e202000867). Our manuscript is focused on describing the details of our method used in the DREAM Challenge.

We also expanded the Discussion section of our manuscript. In addition to summarizing our method and results, we also pointed that gene-gene correlation and spatial correlation are useful information for spatial cell position and gene expression prediction.

The design of figures for a manuscript should be chosen to clearly emphasize the main points of the data. Here, the figures are difficult to interpret because key information and description is missing (e.g., legends are too brief, axis labels are missing, subplot headings are redundant, figure design is inconsistent). In addition, the design and presentation of the figures seem unmotivated and thereby difficult for the reader to see the importance of any particular data or easily draw conclusions from the results.

In addition, consistent colour schemes and designs would help the reader to follow the manuscript and understand the results.

We thank the reviewer for the constructive comments. In the revised manuscript, we improved all figures and legends. We enlarged the font sizes for all figures. In Figure 1, we clarified the whole workflow to display the motivation of our method. In Figure 2, we changed it to display the number of bins with high similarity values (> 0.8) in the similarity matrix for each cell when the number of driver genes is 60, 40, or 20. In Figure 3, we enlarged the font size and added more information in the figure labels. In Figure 4, we enlarged the font size. In figure 5, we changed it to the comparison of the Pearson correlation of predicted gene expression from DistMap (using 84 driver genes) and our method (using 60, 40, or 20 genes).
Competing Interests: No competing interests were disclosed. Close
Report a concern

Views

Reviewer Report 02 Sep 2020

Xianwen Ren, Biomedical Pioneering Innovation Center, Peking University, Beijing, China

Not Approved

https://doi.org/10.5256/f1000research.22477.r69520

Reconstructing the spatial information of single cells from single-cell RNA-seq data is a pivotal question to further release the revolutionary power of the scRNA-seq technology. Here the authors propose a computational method to infer the spatial positions of single cells of Drosophila embryos based on scRNA-seq data and a reference spatial map based on in situ hybridization of 84 driver genes. While the method is finely tuned for this specific scientific question, some concerns exist.

First, the title is misleading. The authors claimed "an unsupervised learning method". This is not valid because this method uses the reference spatial map of Drosophila embryos. Unless the method only uses the scRNA-seq data, it cannot be claimed "an unsupervised learning method".

Second, the current method has only been evaluated on one dataset. More independent evaluations should be added to demonstrate the generalizing capacity of this method.

Overall, this manuscript addresses an important scientific question. But the method needs more justifications and evidence to demonstrate its novelty and power.

Is the rationale for developing the new method (or application) clearly explained?

Yes
Is the description of the method technically sound?

Yes
Are sufficient details provided to allow replication of the method development and its use by others?

Yes
If any results are presented, are all the source data underlying the results available to ensure full reproducibility?

No source data required
Are the conclusions about the method and its performance adequately supported by the findings presented in the article?

Partly

Competing Interests: No competing interests were disclosed.

Reviewer Expertise: Bioinformatics

CITE

Report a concern

Author Response 07 Jun 2021

Zhengqing Ouyang, Department of Biostatistics and Epidemiology, School of Public Health and Health Sciences, University of Massachusetts, Amherst, USA

07 Jun 2021

Author Response

Point-by-point response to the comments of Reviewer 1

Reviewer 1

Reconstructing the spatial information of single cells from single-cell RNA-seq data is a pivotal question to further release ... Continue reading Point-by-point response to the comments of Reviewer 1

Reviewer 1

Reconstructing the spatial information of single cells from single-cell RNA-seq data is a pivotal question to further release the revolutionary power of the scRNA-seq technology. Here the authors propose a computational method to infer the spatial positions of single cells of Drosophila embryos based on scRNA-seq data and a reference spatial map based on in situ hybridization of 84 driver genes. While the method is finely tuned for this specific scientific question, some concerns exist.

1. First, the title is misleading. The authors claimed "an unsupervised learning method". This is not valid because this method uses the reference spatial map of Drosophila embryos. Unless the method only uses the scRNA-seq data, it cannot be claimed "an unsupervised learning method".

We thank the reviewer for the comments. To clarify, we used unsupervised hierarchical clustering for gene selection. Based on the selected genes, we predicted the spatial position of each individual cell. Unsupervised and supervised gene selection approaches are used by the methods in the DREAM Single Cell Transcriptomics Challenge for predicting cell spatial position. Table S2 of the consortium paper (Gene selection for optimal prediction of cell position in tissues from single-cell transcriptomics data, Life Sci Alliance. 2020 Nov; 3(11): e202000867) listed the top performing methods (included ours) using either unsupervised or supervised gene selection.

2.Second, the current method has only been evaluated on one dataset. More independent evaluations should be added to demonstrate the generalizing capacity of this method.

We thank the reviewer for the constructive comments. The results of our method (Zho_team) and other top performing methods have been systematically and independently evaluated on the Drosophila embryo dataset in the DREAM Single Cell Transcriptomics Challenge (https://www.synapse.org/#!Synapse:syn15665609/wiki/583234) and in the consortium paper aforementioned.

Our manuscript, collected in the F1000Research DREAM Challenges Gateway (https://f1000research.com/gateways/dreamchallenges/singlecelltranscriptomics), is focused on describing the details of our method used in the DREAM Challenge. To avoid ambiguity, in the revised manuscript, we changed the title of our manuscript to “Unsupervised gene selection for predicting cell spatial positions in the Drosophila embryo”.

Overall, this manuscript addresses an important scientific question. But the method needs more justifications and evidence to demonstrate its novelty and power.

We thank the reviewer for the comments. Our method is the only one that uses unsupervised hierarchical clustering for gene selection among the top performing methods as described in Table S2 of the DREAM Challenge consortium paper mentioned above. The power has been systematically demonstrated in our manuscript as well as the consortium paper using independent measurements and comprehensive comparisons.
Point-by-point response to the comments of Reviewer 1

Reviewer 1

Reconstructing the spatial information of single cells from single-cell RNA-seq data is a pivotal question to further release the revolutionary power of the scRNA-seq technology. Here the authors propose a computational method to infer the spatial positions of single cells of Drosophila embryos based on scRNA-seq data and a reference spatial map based on in situ hybridization of 84 driver genes. While the method is finely tuned for this specific scientific question, some concerns exist.

1. First, the title is misleading. The authors claimed "an unsupervised learning method". This is not valid because this method uses the reference spatial map of Drosophila embryos. Unless the method only uses the scRNA-seq data, it cannot be claimed "an unsupervised learning method".

We thank the reviewer for the comments. To clarify, we used unsupervised hierarchical clustering for gene selection. Based on the selected genes, we predicted the spatial position of each individual cell. Unsupervised and supervised gene selection approaches are used by the methods in the DREAM Single Cell Transcriptomics Challenge for predicting cell spatial position. Table S2 of the consortium paper (Gene selection for optimal prediction of cell position in tissues from single-cell transcriptomics data, Life Sci Alliance. 2020 Nov; 3(11): e202000867) listed the top performing methods (included ours) using either unsupervised or supervised gene selection.

2.Second, the current method has only been evaluated on one dataset. More independent evaluations should be added to demonstrate the generalizing capacity of this method.

We thank the reviewer for the constructive comments. The results of our method (Zho_team) and other top performing methods have been systematically and independently evaluated on the Drosophila embryo dataset in the DREAM Single Cell Transcriptomics Challenge (https://www.synapse.org/#!Synapse:syn15665609/wiki/583234) and in the consortium paper aforementioned.

Our manuscript, collected in the F1000Research DREAM Challenges Gateway (https://f1000research.com/gateways/dreamchallenges/singlecelltranscriptomics), is focused on describing the details of our method used in the DREAM Challenge. To avoid ambiguity, in the revised manuscript, we changed the title of our manuscript to “Unsupervised gene selection for predicting cell spatial positions in the Drosophila embryo”.

Overall, this manuscript addresses an important scientific question. But the method needs more justifications and evidence to demonstrate its novelty and power.

We thank the reviewer for the comments. Our method is the only one that uses unsupervised hierarchical clustering for gene selection among the top performing methods as described in Table S2 of the DREAM Challenge consortium paper mentioned above. The power has been systematically demonstrated in our manuscript as well as the consortium paper using independent measurements and comprehensive comparisons.
Competing Interests: No competing interests were disclosed. Close
Report a concern
Reviewer Response 17 Jun 2021

Xianwen Ren, Biomedical Pioneering Innovation Center, Peking University, Beijing, China

17 Jun 2021

Reviewer Response

The authors have clarified my concerns.
Competing Interests: NA
The authors have clarified my concerns.
The authors have clarified my concerns.
Competing Interests: NA Close
Report a concern
Respond or Comment

COMMENTS ON THIS REPORT

Author Response 07 Jun 2021

Zhengqing Ouyang, Department of Biostatistics and Epidemiology, School of Public Health and Health Sciences, University of Massachusetts, Amherst, USA

07 Jun 2021

Author Response

Point-by-point response to the comments of Reviewer 1

Reviewer 1

Reconstructing the spatial information of single cells from single-cell RNA-seq data is a pivotal question to further release ... Continue reading Point-by-point response to the comments of Reviewer 1

Reviewer 1

Reconstructing the spatial information of single cells from single-cell RNA-seq data is a pivotal question to further release the revolutionary power of the scRNA-seq technology. Here the authors propose a computational method to infer the spatial positions of single cells of Drosophila embryos based on scRNA-seq data and a reference spatial map based on in situ hybridization of 84 driver genes. While the method is finely tuned for this specific scientific question, some concerns exist.

1. First, the title is misleading. The authors claimed "an unsupervised learning method". This is not valid because this method uses the reference spatial map of Drosophila embryos. Unless the method only uses the scRNA-seq data, it cannot be claimed "an unsupervised learning method".

We thank the reviewer for the comments. To clarify, we used unsupervised hierarchical clustering for gene selection. Based on the selected genes, we predicted the spatial position of each individual cell. Unsupervised and supervised gene selection approaches are used by the methods in the DREAM Single Cell Transcriptomics Challenge for predicting cell spatial position. Table S2 of the consortium paper (Gene selection for optimal prediction of cell position in tissues from single-cell transcriptomics data, Life Sci Alliance. 2020 Nov; 3(11): e202000867) listed the top performing methods (included ours) using either unsupervised or supervised gene selection.

2.Second, the current method has only been evaluated on one dataset. More independent evaluations should be added to demonstrate the generalizing capacity of this method.

We thank the reviewer for the constructive comments. The results of our method (Zho_team) and other top performing methods have been systematically and independently evaluated on the Drosophila embryo dataset in the DREAM Single Cell Transcriptomics Challenge (https://www.synapse.org/#!Synapse:syn15665609/wiki/583234) and in the consortium paper aforementioned.

Our manuscript, collected in the F1000Research DREAM Challenges Gateway (https://f1000research.com/gateways/dreamchallenges/singlecelltranscriptomics), is focused on describing the details of our method used in the DREAM Challenge. To avoid ambiguity, in the revised manuscript, we changed the title of our manuscript to “Unsupervised gene selection for predicting cell spatial positions in the Drosophila embryo”.

Overall, this manuscript addresses an important scientific question. But the method needs more justifications and evidence to demonstrate its novelty and power.

We thank the reviewer for the comments. Our method is the only one that uses unsupervised hierarchical clustering for gene selection among the top performing methods as described in Table S2 of the DREAM Challenge consortium paper mentioned above. The power has been systematically demonstrated in our manuscript as well as the consortium paper using independent measurements and comprehensive comparisons.
Point-by-point response to the comments of Reviewer 1

Reviewer 1

Reconstructing the spatial information of single cells from single-cell RNA-seq data is a pivotal question to further release the revolutionary power of the scRNA-seq technology. Here the authors propose a computational method to infer the spatial positions of single cells of Drosophila embryos based on scRNA-seq data and a reference spatial map based on in situ hybridization of 84 driver genes. While the method is finely tuned for this specific scientific question, some concerns exist.

1. First, the title is misleading. The authors claimed "an unsupervised learning method". This is not valid because this method uses the reference spatial map of Drosophila embryos. Unless the method only uses the scRNA-seq data, it cannot be claimed "an unsupervised learning method".

We thank the reviewer for the comments. To clarify, we used unsupervised hierarchical clustering for gene selection. Based on the selected genes, we predicted the spatial position of each individual cell. Unsupervised and supervised gene selection approaches are used by the methods in the DREAM Single Cell Transcriptomics Challenge for predicting cell spatial position. Table S2 of the consortium paper (Gene selection for optimal prediction of cell position in tissues from single-cell transcriptomics data, Life Sci Alliance. 2020 Nov; 3(11): e202000867) listed the top performing methods (included ours) using either unsupervised or supervised gene selection.

2.Second, the current method has only been evaluated on one dataset. More independent evaluations should be added to demonstrate the generalizing capacity of this method.

We thank the reviewer for the constructive comments. The results of our method (Zho_team) and other top performing methods have been systematically and independently evaluated on the Drosophila embryo dataset in the DREAM Single Cell Transcriptomics Challenge (https://www.synapse.org/#!Synapse:syn15665609/wiki/583234) and in the consortium paper aforementioned.

Our manuscript, collected in the F1000Research DREAM Challenges Gateway (https://f1000research.com/gateways/dreamchallenges/singlecelltranscriptomics), is focused on describing the details of our method used in the DREAM Challenge. To avoid ambiguity, in the revised manuscript, we changed the title of our manuscript to “Unsupervised gene selection for predicting cell spatial positions in the Drosophila embryo”.

Overall, this manuscript addresses an important scientific question. But the method needs more justifications and evidence to demonstrate its novelty and power.

We thank the reviewer for the comments. Our method is the only one that uses unsupervised hierarchical clustering for gene selection among the top performing methods as described in Table S2 of the DREAM Challenge consortium paper mentioned above. The power has been systematically demonstrated in our manuscript as well as the consortium paper using independent measurements and comprehensive comparisons.
Competing Interests: No competing interests were disclosed. Close
Report a concern
Reviewer Response 17 Jun 2021

Xianwen Ren, Biomedical Pioneering Innovation Center, Peking University, Beijing, China

17 Jun 2021

Reviewer Response

The authors have clarified my concerns.
Competing Interests: NA
The authors have clarified my concerns.
The authors have clarified my concerns.
Competing Interests: NA Close
Report a concern

Comments on this article Comments (0)

Version 2

VERSION 2 PUBLISHED 19 Feb 2020

Open Peer Review

Reviewer Status

Reviewer Reports

	Invited Reviewers
	1	2	3	4	5
Version 2 (revision) 07 Jun 21	read		read	read	read
Version 1 19 Feb 20	read	read

Xianwen Ren, Peking University, Beijing, China
Mark S. Cembrowski, University of British Columbia, Vancouver, Canada

Larissa Kraus, University of British Columbia, Vancouver, Canada
Yen-Chung Chen, New York University, New York, USA
Saumitra Dey Choudhury, All India Institute of Medical Sciences, New Delhi, India
Komal Kumar Bollepogu Raja, Baylor college of medicine, Houston, USA

Comments on this article

All Comments(0)

Add a comment

Browse by related subjects

Back to all reports

Reviewer Report

8 Views

28 Jun 2024 | for Version 2

Komal Kumar Bollepogu Raja, Baylor college of medicine, Houston, TX, USA

8 Views Cite this report Responses(0)

Approved With Reservations

Is the rationale for developing the new method (or application) clearly explained?

Yes
Is the description of the method technically sound?

Yes
Are sufficient details provided to allow replication of the method development and its use by others?

Yes
If any results are presented, are all the source data underlying the results available to ensure full reproducibility?

Yes
Are the conclusions about the method and its performance adequately supported by the findings presented in the article?

Partly

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

Single cell omics, Genomics and Bioinformatics

Respond to this report

Responses (0)

Back to all reports

Reviewer Report

10 Views

28 Jun 2024 | for Version 2

Saumitra Dey Choudhury, All India Institute of Medical Sciences, New Delhi, India

10 Views Cite this report Responses(0)

Not Approved

Is the rationale for developing the new method (or application) clearly explained?

Yes
Is the description of the method technically sound?

Yes
Are sufficient details provided to allow replication of the method development and its use by others?

Yes
If any results are presented, are all the source data underlying the results available to ensure full reproducibility?

No source data required
Are the conclusions about the method and its performance adequately supported by the findings presented in the article?

No

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

Drosophila Neurogenetics, Single-cell RNA Transcriptomics, Confocal and Super-resolution Microscopy

Respond to this report

Responses (0)

Back to all reports

Reviewer Report

9 Views

27 Jun 2024 | for Version 2

Yen-Chung Chen, New York University, New York, New York, USA

9 Views Cite this report Responses(0)

Not Approved

Is the rationale for developing the new method (or application) clearly explained?

Yes
Is the description of the method technically sound?

Partly
Are sufficient details provided to allow replication of the method development and its use by others?

Yes
If any results are presented, are all the source data underlying the results available to ensure full reproducibility?

No source data required
Are the conclusions about the method and its performance adequately supported by the findings presented in the article?

Partly

References

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

Genomics, developmental biology, biostatistics

Respond to this report

Responses (0)

Back to all reports

Reviewer Report

10 Views

17 Jun 2021 | for Version 2

Xianwen Ren, Biomedical Pioneering Innovation Center, Peking University, Beijing, China

10 Views Cite this report Responses(0)

Approved

The authors have clarified my concerns.

Competing Interests

No competing interests were disclosed.

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Respond to this report

Responses (0)

Back to all reports

Reviewer Report

29 Views

23 Sep 2020 | for Version 1

Mark S. Cembrowski, Dept. of Cellular and Physiological Sciences, University of British Columbia, Vancouver, BC, Canada

Larissa Kraus, Department of Cellular and Physiological Sciences, University of British Columbia, Vancouver, BC, Canada

29 Views Cite this report Responses(1)

Not Approved

Is the rationale for developing the new method (or application) clearly explained?

Yes
Is the description of the method technically sound?

Yes
Are sufficient details provided to allow replication of the method development and its use by others?

Partly
If any results are presented, are all the source data underlying the results available to ensure full reproducibility?

Yes
Are the conclusions about the method and its performance adequately supported by the findings presented in the article?

No

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

Transcriptomics

Respond to this report

Responses (1)

Author Response

07 Jun 2021

Zhengqing Ouyang, Department of Biostatistics and Epidemiology, School of Public Health and Health Sciences, University of Massachusetts, Amherst, USA

Point-by-point response to the comments of Reviewer 2

Reviewer 2

In this manuscript, the authors describe an approach to infer spatial gene expression using scRNA-seq data. The authors aimed to identify spatial expression of cell clusters using 20, 40 or 60 marker genes, and compared to the “gold standard” Distmap method using 84 marker genes.

The authors begin to explain the rationale of the manuscript by describing the concept of the DREAM challenge, which aims to introduce novel computational techniques to map spatial gene expression using transcriptomic data.
Since the DREAM challenge aims to implement novel methods, it is imperative to introduce and carefully compare common methods with the novel proposed approach, which was underdeveloped by the authors. As an example, the “gold standard” method is mentioned several times, though it is not clear what exactly the “gold standard” is or how well-controlled it is. As another key example, a discussion of any kind is completely missing.

We thank the reviewer for the constructive comments. The “gold standard” is generated by DistMap, which is the best available single cell position reference used in the DREAM Challenge. DistMap determines single cell position by searching the maximum similarity for cells and spatial bins using the in situ expression of 84 driver genes in the Berkeley Drosophila Transcription Network Project (BDNTP). It is shown that the 84 driver gens are sufficient to uniquely and individually label most of the 1,297 cells (Karaiskos, et al. 2017). Thus, it is used in the DREAM Challenge to assess the performance of methods using smaller numbers (60, 40, and 20) of driver genes. In the revised manuscript, we changed “gold standard” to “silver standard”, as the cell positions generated by DistMap are not experimentally validated. We added the description of the DistMap “silver standard” in the Method section.

We also added more introductions to the common methods for the cell position prediction from single cell transcriptome data in the Introduction section. We described the limitation of current methods and explain the motivation to study how the driver gene selection affecting the prediction of cell spatial position.

The comparisons of our methods (Zho_team) and other top performing methods have been described in the DREAM Single Cell Transcriptomics Challenge consortium paper (Gene selection for optimal prediction of cell position in tissues from single-cell transcriptomics data, Life Sci Alliance. 2020 Nov; 3(11): e202000867). Our manuscript is focused on describing the details of our method used in the DREAM Challenge.

We also expanded the Discussion section of our manuscript. In addition to summarizing our method and results, we also pointed that gene-gene correlation and spatial correlation are useful information for spatial cell position and gene expression prediction.

The design of figures for a manuscript should be chosen to clearly emphasize the main points of the data. Here, the figures are difficult to interpret because key information and description is missing (e.g., legends are too brief, axis labels are missing, subplot headings are redundant, figure design is inconsistent). In addition, the design and presentation of the figures seem unmotivated and thereby difficult for the reader to see the importance of any particular data or easily draw conclusions from the results.

In addition, consistent colour schemes and designs would help the reader to follow the manuscript and understand the results.

We thank the reviewer for the constructive comments. In the revised manuscript, we improved all figures and legends. We enlarged the font sizes for all figures. In Figure 1, we clarified the whole workflow to display the motivation of our method. In Figure 2, we changed it to display the number of bins with high similarity values (> 0.8) in the similarity matrix for each cell when the number of driver genes is 60, 40, or 20. In Figure 3, we enlarged the font size and added more information in the figure labels. In Figure 4, we enlarged the font size. In figure 5, we changed it to the comparison of the Pearson correlation of predicted gene expression from DistMap (using 84 driver genes) and our method (using 60, 40, or 20 genes).

View more View less

Competing Interests

No competing interests were disclosed.

Back to all reports

Reviewer Report

35 Views

02 Sep 2020 | for Version 1

Xianwen Ren, Biomedical Pioneering Innovation Center, Peking University, Beijing, China

35 Views Cite this report Responses(2)

Not Approved

Is the rationale for developing the new method (or application) clearly explained?

Yes
Is the description of the method technically sound?

Yes
Are sufficient details provided to allow replication of the method development and its use by others?

Yes
If any results are presented, are all the source data underlying the results available to ensure full reproducibility?

No source data required
Are the conclusions about the method and its performance adequately supported by the findings presented in the article?

Partly

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

Bioinformatics

Respond to this report

Responses (2)

Author Response

07 Jun 2021

Zhengqing Ouyang, Department of Biostatistics and Epidemiology, School of Public Health and Health Sciences, University of Massachusetts, Amherst, USA

Point-by-point response to the comments of Reviewer 1

Reviewer 1

Reconstructing the spatial information of single cells from single-cell RNA-seq data is a pivotal question to further release the revolutionary power of the scRNA-seq technology. Here the authors propose a computational method to infer the spatial positions of single cells of Drosophila embryos based on scRNA-seq data and a reference spatial map based on in situ hybridization of 84 driver genes. While the method is finely tuned for this specific scientific question, some concerns exist.

1. First, the title is misleading. The authors claimed "an unsupervised learning method". This is not valid because this method uses the reference spatial map of Drosophila embryos. Unless the method only uses the scRNA-seq data, it cannot be claimed "an unsupervised learning method".

We thank the reviewer for the comments. To clarify, we used unsupervised hierarchical clustering for gene selection. Based on the selected genes, we predicted the spatial position of each individual cell. Unsupervised and supervised gene selection approaches are used by the methods in the DREAM Single Cell Transcriptomics Challenge for predicting cell spatial position. Table S2 of the consortium paper (Gene selection for optimal prediction of cell position in tissues from single-cell transcriptomics data, Life Sci Alliance. 2020 Nov; 3(11): e202000867) listed the top performing methods (included ours) using either unsupervised or supervised gene selection.

2.Second, the current method has only been evaluated on one dataset. More independent evaluations should be added to demonstrate the generalizing capacity of this method.

We thank the reviewer for the constructive comments. The results of our method (Zho_team) and other top performing methods have been systematically and independently evaluated on the Drosophila embryo dataset in the DREAM Single Cell Transcriptomics Challenge (https://www.synapse.org/#!Synapse:syn15665609/wiki/583234) and in the consortium paper aforementioned.

Our manuscript, collected in the F1000Research DREAM Challenges Gateway (https://f1000research.com/gateways/dreamchallenges/singlecelltranscriptomics), is focused on describing the details of our method used in the DREAM Challenge. To avoid ambiguity, in the revised manuscript, we changed the title of our manuscript to “Unsupervised gene selection for predicting cell spatial positions in the Drosophila embryo”.

Overall, this manuscript addresses an important scientific question. But the method needs more justifications and evidence to demonstrate its novelty and power.

We thank the reviewer for the comments. Our method is the only one that uses unsupervised hierarchical clustering for gene selection among the top performing methods as described in Table S2 of the DREAM Challenge consortium paper mentioned above. The power has been systematically demonstrated in our manuscript as well as the consortium paper using independent measurements and comprehensive comparisons.

View more View less

Competing Interests

No competing interests were disclosed.

Alongside their report, reviewers assign a status to the article:

Approved - the paper is scientifically sound in its current form and only minor, if any, improvements are suggested

Approved with reservations - A number of small changes, sometimes more significant revisions are required to address specific details and improve the papers academic merit.

Not approved - fundamental flaws in the paper seriously undermine the findings and conclusions

[1] 1. Farrell JA, Wang Y, Riesenfeld SJ, et al.: Single-cell reconstruction of developmental trajectories during zebrafish embryogenesis. Science. 2018; 360(6392): eaar3131. PubMed Abstract | Publisher Full Text | Free Full Text

[2] 2. Butler A, Hoffman P, Smibert P, et al.: Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat Biotechnol. 2018; 36(5): 411–420. PubMed Abstract | Publisher Full Text | Free Full Text

[3] 3. Colomé-Tatché M, Theis FJ: Statistical single cell multi-omics integration. Curr Opin Syst Biol. 2018; 7: 54–59. Publisher Full Text

[4] 4. Gu C, Liu S, Wu Q, et al.: Integrative single-cell analysis of transcriptome, DNA methylome and chromatin accessibility in mouse oocytes. Cell Res. 2019; 29(2): 110–123. PubMed Abstract | Publisher Full Text | Free Full Text

[5] 5. Bian S, Hou Y, Zhou X, et al.: Single-cell multiomics sequencing and analyses of human colorectal cancer. Science. 2018; 362(6418): 1060–1063. PubMed Abstract | Publisher Full Text

[6] 6. Kulkarni A, Anderson AG, Merullo Dp, et al.: Beyond bulk: a review of single cell transcriptomics methodologies and applications. Curr Opin Biotechnol. 2019; 58: 129–136. PubMed Abstract | Publisher Full Text | Free Full Text

[7] 7. Savulescu AF, Jacobs C, Negishi Y, et al.: Pinpointing cell identity in time and space. Front Mol Biosci. 2020; 7: 209. PubMed Abstract | Publisher Full Text | Free Full Text

[8] 8. Stuart T, Satija R: Integrative single-cell analysis. Nat Rev Genet. 2019; 20(5): 257–272. PubMed Abstract | Publisher Full Text

[9] 9. Mayr U, Serra D, Liberali P: Exploring single cells in space and time during tissue development, homeostasis and regeneration. Development. 2019; 146(12): dev176727. PubMed Abstract | Publisher Full Text

[10] 10. Karaiskos N, Wahle P, Alles J, et al.: The Drosophila embryo at single-cell transcriptome resolution. Science. 2017; 358(6360): 194–199. PubMed Abstract | Publisher Full Text

[11] 11. Satija R, Farrell JA, Gennert D, et al.: Spatial reconstruction of single-cell gene expression data. Nat Biotechnol. 2015; 33(5): 495–502. PubMed Abstract | Publisher Full Text | Free Full Text

[12] 12. Achim K, Pettit JB, Saraiva LR, et al.: High-throughput spatial mapping of single-cell RNA-seq data to tissue of origin. Nat Biotechnol. 2015; 33(5): 503–9. PubMed Abstract | Publisher Full Text

[13] 13. Moor AE, Harnik Y, Ben-Moshe S, et al.: Spatial reconstruction of single enterocytes uncovers broad zonation along the intestinal villus axis. Cell. 2018; 175(4): 1156–1167.e15. PubMed Abstract | Publisher Full Text

[14] 14. Nitzan M, Karaiskos N, Friedman N, et al.: Charting a tissue from single-cell transcriptomes. bioRxiv. 2018; 456350. Publisher Full Text

[15] 15. Nitzan M, Karaiskos N, Friedman N, et al.: Gene expression cartography. Nature. 2019; 576(7785): 132–137. PubMed Abstract | Publisher Full Text

[16] 16. Shah S, Lubeck E, Zhou W, et al.: In situ transcription profiling of single cells reveals spatial organization of cells in the mouse hippocampus. Neuron. 2016; 92(2): 342–357. PubMed Abstract | Publisher Full Text | Free Full Text

[17] 17. Zhu Q, Shah S, Dries R, et al.: Identification of spatially associated subpopulations by combining scRNAseq and sequential fluorescence in situ hybridization data. Nat Biotechnol. 2018; 36: 1183–1190. PubMed Abstract | Publisher Full Text | Free Full Text

[18] 18. Moffitt JR, Bambah-Mukku D, Eichhorn SW, et al.: Molecular, spatial, and functional single-cell profiling of the hypothalamic preoptic region. Science. 2018; 362(6416): eaau5324. PubMed Abstract | Publisher Full Text | Free Full Text

[19] 19. Halpern KB, Shenhav R, Matcovitch-Natan O, et al.: Single-cell spatial reconstruction reveals global division of labour in the mammalian liver. Nature. 2017; 542(7641): 352–356. PubMed Abstract | Publisher Full Text | Free Full Text

[20] 20. Tanevski J, Nguyen T, Truong B, et al.: Gene selection for optimal prediction of cell position in tissues from single-cell transcriptomics data. Life Sci Alliance. 2020; 3(11): e202000867. PubMed Abstract | Publisher Full Text | Free Full Text

[21] 21. Macosko EZ, Basu A, Satija R, et al.: Highly Parallel Genome-wide Expression Profiling of Individual Cells Using Nanoliter Droplets. Cell. 2015; 161(5): 1202–1214. PubMed Abstract | Publisher Full Text | Free Full Text

[22] 22. Charrad M, Ghazzali N, Boiteau V, et al.: NbClust: An R Package for Determining the Relevant Number of Clusters in a Data Set. J Stat Softw. 2014; 61(6): 1–36. Publisher Full Text

Unsupervised gene selection for predicting cell spatial positions in the Drosophila embryo

Abstract

Keywords

Revised Amendments from Version 1

Introduction

Methods

Dataset

Gene selection

Figure 1. Workflow of our spatial position prediction method for scRNA-seq with in situ reference gene panel.

Binarization of scRNA-seq data

Compute the similarity matrix between cells and bins

Select candidate the cell positions based on the similarity matrix

Figure 2. Compare the number of high similarity values (> 0.8) in the similarity matrix in each bin for sub-challenge 1, 2, and 3.

Silhouette score to determine the number of clusters of bins

Predict cell positions based on clustering of bins

Predict spatial gene expression based on the similarity matrix

Performance evaluation

Results

Select driver genes

Figure 3.

Evaluate the predicted position and spatial gene expression

Figure 4.

Figure 5. Comparing Pearson correlation of predicted spatial gene expression and in situ spatial gene expression for 84 genes between DistMap (84 genes) with our method using (a) 60 genes in subchallenge 1; (b) 40 genes in sub-challenge 2; and (c) 20 genes in sub-challenge 3.

Discussion

Data availability

Software availability

Author information

Acknowledgements

References

Comments on this article Comments (0)

Open Peer Review

Comments on this article Comments (0)

Open Peer Review

Reviewer Status

Reviewer Reports

Comments on this article

Browse by related subjects

Competing Interests Policy

Stay Updated