SemOpenAlex |

SemOpenAlex

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4205656760> ?p ?o ?g. }

Showing items 1 to 61 of 61 with 100 items per page.

W4205656760 abstract "Article Figures and data Abstract eLife digest Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Multidimensional landscapes of regulatory genes in neuronal phenotypes at whole-brain levels in the vertebrate remain elusive. We generated single-cell transcriptomes of ~67,000 region- and neurotransmitter/neuromodulator-identifiable cells from larval zebrafish brains. Hierarchical clustering based on effector gene profiles (‘terminal features’) distinguished major brain cell types. Sister clusters at hierarchical termini displayed similar terminal features. It was further verified by a population-level statistical method. Intriguingly, glutamatergic/GABAergic sister clusters mostly expressed distinct transcription factor (TF) profiles (‘convergent pattern’), whereas neuromodulator-type sister clusters predominantly expressed the same TF profiles (‘matched pattern’). Interestingly, glutamatergic/GABAergic clusters with similar TF profiles could also display different terminal features (‘divergent pattern’). It led us to identify a library of RNA-binding proteins that differentially marked divergent pair clusters, suggesting the post-transcriptional regulation of neuron diversification. Thus, our findings reveal multidimensional landscapes of transcriptional and post-transcriptional regulators in whole-brain neuronal phenotypes in the zebrafish brain. eLife digest The brain harbors an astounding diversity of interconnected cells. Each cell contains the same basic set of genetic instructions, but only a fraction of the genome is used in each cell to assemble proteins. This selective gene expression gives rise to each cell’s characteristic properties, such as their shape and location, or whether they can activate or inhibit neighbouring cells. How these defining features are encoded on a genetic level and selectively activated in cells to produce such diversity in the brain is not fully understood. One way to study gene expression in single cells involves profiling the transcriptome, the full range of intermediary RNA molecules a cell produces from its genes to make proteins. Zhang et al. used transcriptome profiling to better understand how thousands of regulatory genes encoding regulatory proteins called transcription factors create different types of neurons in the zebrafish brain, which is similar to but much simpler than the human brain. To do so, they analysed transcriptome data extracted from cell populations located in specific brain regions and displaying different properties. Zhang et al. identified distinct clusters of neurons in the larval zebrafish brain. Mathematical models then analysed the transcriptome profiles of these neuronal clusters with characteristic features. They revealed that neurons with similar characteristics did not necessarily share the same transcription factors. In other words, distinct sets of transcription factors gave rise to the same types of cells. Zhang et al. described this observation as a ‘convergent’ pattern. On the contrary, some neurons with dissimilar features expressed the same sorts of transcription factors, suggesting a ‘divergent’ developmental pattern also exists. In summary, this work sheds light on variable gene expression patterns akin to design principles that shape neuronal diversity in the brain. It gives a new appreciation of how neuronal subtypes result from a complex set of regulatory factors controlling gene expression. Introduction The vertebrate brain harbors highly diverse neuronal types that are specifically interconnected to form functional circuits (Kepecs and Fishell, 2014; Armañanzas and Ascoli, 2015; Moffitt et al., 2018). The brain comprises conserved major cell types (neurons, progenitors, glia, endothelial cells, etc.) (Marques et al., 2016; Saunders et al., 2018; Hodge et al., 2019). Previous studies have indicated that neuronal types could be well characterized by their features, including electrophysiological properties, neurotransmitter/modulator identity, synaptic connectivity, brain region identity, and cellular morphology (Zeisel et al., 2015; Pandey et al., 2018; Zeisel et al., 2018). How these neuronal features are molecularly encoded at the whole-brain level is an important issue in neuroscience. With the advent of single-cell RNA-sequencing (scRNA-seq) technology, recent studies have begun to characterize neuronal diversity by analyzing single-cell transcriptomes of large populations in the whole brain or specific brain regions of mice and humans (Darmanis et al., 2015; Lake et al., 2016; Poulin et al., 2016; Tasic et al., 2016; Chen et al., 2017; Tasic et al., 2018). These studies have provided compelling evidence that individual neuronal types could be well characterized by their transcriptomes (Poulin et al., 2016; Tasic et al., 2018; Sugino et al., 2019). For instance, scRNA-seq analyses of half a million cells from the mouse brain showed that neurons could be classified by genes related to neuronal connectivity, synaptic transmission, and membrane conductance (Zeisel et al., 2018). Besides, the scRNA-seq analysis of nearly 200 genetically marked mouse neuronal populations also showed that neurons could be classified by the expression level of various transcription factors (TFs), ion channels, synaptic proteins, and cell adhesion molecules (Sugino et al., 2019). These studies provided extensive information on the molecules that could be used to define neuronal types. Generally, these molecules could be sub-divided into two primary categories: regulatory genes and effector genes. Regulatory genes encode proteins involved in gene transcription and translation (e.g., TFs and post-transcriptional regulators), while effector genes encode proteins serving specific neuronal terminal features (e.g., synaptic proteins, ion channels, transporters, and receptors) (Kratsios et al., 2015; Zeisel et al., 2015; Paul et al., 2017, Zeisel et al., 2018; Reilly et al., 2020). Regulatory genes are critical for establishing and maintaining effector gene profiles that resulted in diverse neuronal types. Interestingly, distinct TFs can determine neuronal types with similar neurotransmitter identities in the Caenorhabditis elegans, arguing for phenotypic convergence (Serrano-Saiz et al., 2013; Gendrel et al., 2016; Hobert and Kratsios, 2019). Remarkably, this phenotypic convergence has also been reported in the Drosophila’s optic lobe (Konstantinides et al., 2018). However, the multidimensional landscapes of regulatory genes in vertebrate whole-brain neuronal phenotypes remain elusive. The larval zebrafish brain comprises only about 100,000 cells, thus providing an outstanding vertebrate model for studying cell diversity within the entire brain, using single-cell transcriptome analysis with full cell coverage by the 10× Genomics Platform. Previous scRNA analysis of the zebrafish nervous system has elegantly demonstrated the temporal dynamics of brain cell development (Raj et al., 2018; Raj et al., 2020). In this study, we generated the multidimensional landscape of regulator genes in effector gene-based neuronal phenotypes (terminal features) at the whole-brain level by combining single-cell transcriptome data obtained from the whole brain, specific brain regions, as well as neurotransmitter- and neuromodulator-defined neuronal populations. We found that, at the transcriptional and post-transcriptional levels, glutamatergic/GABAergic neurons with the same terminal features could express different TF profiles, while those with different terminal features could express the same TF but different RNA-binding protein (RBP) profiles. In contrast, neuromodulator-type neurons that display particular terminal features expressed unique TF profiles. Thus, our findings reveal multidimensional landscapes of transcriptional and post-transcriptional regulators in the whole zebrafish brain. Results Molecular classification of whole-brain cells in zebrafish To uncover the transcriptomic profiles of diverse cell types with regional identity in the larval zebrafish whole brain at single-cell resolution, we dissected and dissociated cells from the whole brain (n = 4), four specific brain regions (n = 2 each), including the forebrain (Fore), optic tectum (OT), hindbrain (Hind), and the region underneath the optic tectum (sub-OT) in the 8 days post fertilization (dpf) zebrafish (Figure 1—figure supplement 1A). We performed scRNA-seq of these cells using the 10× Genomics Chromium 3’ v2 platform. The libraries were sequenced to a mean depth of 126,651 reads per library, with a median of 1891 UMI and 866 genes per cell (Figure 1—source data 1). Reproducibility of transcription analysis was shown by the finding that replicates of whole-brain samples and individual brain regions were primarily overlapped in the t-distributed stochastic neighbor embedding (t-SNE) plots (Figure 1—figure supplement 1B-C). We obtained the transcriptomes of 65,253 cells and selected 45,746 cells for further analysis after filtering out low-quality data (Materials and methods). Then aggregated all cells into 68 clusters using high-variance genes (n = 1402) in t-SNE plot (Figure 1A). The Jaccard index-based analysis showed that most clusters had a high Jaccard index greater than 0.6, indicating the robustness of these clusters (details in Materials and methods) (Tang et al., 2020) except for a few neuronal clusters from sub-OT (Clusters 52, 66), and non-neuronal clusters (neuroprogenitors: Clusters 37, 56; radial astrocytes: 65, Figure 1—figure supplement 1E). Figure 1 with 2 supplements see all Download asset Open asset Molecular classification of whole-brain cells in larval zebrafish. (A) The t-distributed stochastic neighbor embedding (t-SNE) plot of 45,746 single-cell transcriptomes pooled from whole brains (n = 4) and four different individual brain regions (n = 2 each). The pooled cells were aggregated into 68 clusters, marked by a number. Each color-coded the major cell type as F. (B) The schematic showing different samples separately examined by single-cell RNA-sequencing on 10× Genomics Drop-seq platform: whole brain (n = 4), forebrain (Fore, n = 2), optic tectum (OT, n = 2), hindbrain (Hind, n = 2), and the region underneath the optic tectum (sub-OT, n = 2). OB: olfactory bulb; Tel: telencephalon; OT: optic tectum; Th: thalamus; H: hypothalamus; Pit: pituitary; Ce: cerebellum; MO: medulla oblongata. (C) Venn plots showing the differentially expressed genes in four major cell types identified by cell-type marker genes (vglut+, glutamatergic neurons, Glu; gad1b+, GABAergic neurons, Gaba; pcna+, neuroprogenitors, P; cx43+, radial astrocytes, R) in three brain regions (Fore, Hind, and OT). Commonly expressed genes in all cell types for a given brain region were identified as region-specific genes: six for forebrain (Fore), one for optic tectum (OT), and one for hindbrain (Hind), with genes listed below. (D) Dot plot showing the expression levels of region-specific marker genes in four major cell types (colored circles as C) in three brain regions. The gray level represents the average expression; dot size represents the percentage of cells expressing the marker genes. (E) Lawson-Hanson algorithm for non-negative least squares (NNLS) analysis showed cell clusters of Fore, OT, and Hind exhibited a high correlation with their counterparts of the juvenile zebrafish. Degree of correlation in marker genes is coded by the gray level and size of circle. (F) The dendrogram for the taxonomy of 68 identified clusters based on effector gene profiles (n = 1099). Main branches of neuronal and non-neuronal cells were classified into six branches (red dashed line) that include: I, cerebellum and habenula (hb); IIa, glutamatergic neurons (Glu); IIb, inhibitory neurons (Gaba); III, neuroprogenitors (P); IV, radial astrocytes (R); V, others, including microglia, endothelial cells, and oligodendrocytes. The colored dots and squares below indicate their regional origins and neurotransmitter-type, respectively. Figure 1—source data 1 Bioinformatics processing of raw reads of single-cell samples. https://cdn.elifesciences.org/articles/68224/elife-68224-fig1-data1-v2.xlsx Download elife-68224-fig1-data1-v2.xlsx Figure 1—source data 2 The annotation of 68 clusters of whole-brain sample. https://cdn.elifesciences.org/articles/68224/elife-68224-fig1-data2-v2.xlsx Download elife-68224-fig1-data2-v2.xlsx Figure 1—source data 3 The regional origins and neurotransmitter-type annotation of each whole-brain cluster with well-known markers. https://cdn.elifesciences.org/articles/68224/elife-68224-fig1-data3-v2.xlsx Download elife-68224-fig1-data3-v2.xlsx Figure 1—source data 4 Top 20 marker genes of whole-brain larval zebrafish 68 clusters. https://cdn.elifesciences.org/articles/68224/elife-68224-fig1-data4-v2.xlsx Download elife-68224-fig1-data4-v2.xlsx Figure 1—source data 5 Marker genes of major six cell type in whole brain. https://cdn.elifesciences.org/articles/68224/elife-68224-fig1-data5-v2.xlsx Download elife-68224-fig1-data5-v2.xlsx Each cluster was annotated according to cell-type-specific marker genes from the literature or ZFIN (Zebrafish Information Network) database (Figure 1—source data 2). To assign the brain region identity for each cluster, cells from each of the four specific regions (Fore, OT, sub-OT, Hind) were found to cover multiple but non-overlapping clusters and all 68 clusters could be assigned with their brain region origins (Figure 1—figure supplement 2A-B, and Figure 1—source data 3). Furthermore, to identify potential brain region-specific markers that exist in all cell types, we identified all genes that were differentially expressed in each region for all major cell types (vglut+, glutamatergic neurons; gad+, inhibitory neurons; pcna+, neuroprogenitors; cx43+, radial astrocytes). The differentially expressed genes shared by all cell types were considered to represent the targeted region-specific markers (Figure 1C). We indeed identified a small set of genes specific to each brain region independent of cell type. For instance, foxg1a, en2a, and hoxb3a were explicitly expressed in all cell types of the Fore, OT, and Hind, respectively (Figure 1D). Interestingly, these brain region-specific genes also exhibited a conserved region-specific expression pattern in the mouse brain (Hanks et al., 1995, Manzanares et al., 2001, Kumamoto and Hanashima, 2017). We found no specific gene for the sub-OT, probably due to the diverse brain structures in this region. These region-specific genes, which may be involved in forming regional identity during brain development, could be used to study region-specific neuronal connectivity and function. To assign the neurotransmitter/modulator identity for each cluster, we used the marker genes that were specific to primary neurotransmitter/modulator phenotypes, including slc17a6b (glutamatergic), gad1b (GABAergic), slc6a5 (glycinergic), th (dopaminergic [DA]), tph2 (serotonergic [5-HT]), and chata (cholinergic [ChAT]) (Figure 1—figure supplement 2C, Figure 1—source data 3). The ratio of glutamatergic to GABAergic neurons was the highest in the forebrain and lowest in the hindbrain, indicating that glutamatergic neurons predominantly belonged to the forebrain, whereas glycinergic neurons mainly resided in the hindbrain (Figure 1—figure supplement 2D). These regional patterns of neurons expressing different neurotransmitter types were validated using the transgenic fishlines: Tg (vglut2a:loxp-DsRed-loxp-GFP), Tg (gad1b:EGFP), and Tg (glyT2:GFP), each exhibiting distinct labeling of glutamatergic, GABAergic, and glycinergic neurons, respectively (Figure 1—figure supplement 2E). Moreover, the Lawson-Hanson algorithm for non-negative least squares (NNLS) analysis using cluster-specific marker genes (top 20, Figure 1—source data 4) showed that these 68 clusters exhibited a high overlap with their counterparts in the juvenile zebrafish brain recently reported (Raj et al., 2018). Meanwhile, clusters with different regional origins or cell types also exhibited a high correlation with their counterparts in the juvenile zebrafish (Figure 1E and Figure 1—figure supplement 2F). Thus, our analysis indicated that the brain at 8 dpf mostly represented cellular diversity in the juvenile brain. Furthermore, Gene Ontology (GO) analysis of 1402 variable genes used for the classification of all 68 clusters showed that the majority of these genes (78.4%, n = 1099) were effector genes, which could be classified as neuropeptides, receptors, transporters, ion channels, synaptic proteins, and cell adhesion molecules (Figure 1—figure supplement 2G). This result suggests the importance of effector gene profiles in brain cell classification, which has also been appreciated by previous studies in different species (Paul et al., 2017, Hodge et al., 2019). We thus generated the hierarchical classification of all 68 cell clusters using the profiles of these 1099 effector genes. All 68 clusters were first segregated into two groups, neuronal cells (48 clusters, 37,880 cells) and non-neuronal cells (20 clusters, 7866 cells) (Figure 1F). Among non-neuronal cells, oligodendrocytes (Clusters 40, 42; olig2+, sox10+), microglia (Cluster 55; apoeb+, mpeg1.1+), endothelial cells (Clusters 48, 57; fxyd1+, rbp4+), erythrocytes (Cluster 67; hbbe1.2+, hbae5+), radial astrocytes (Clusters 21, 46, 65; cx43+, glua+), and neuroprogenitors (Clusters 8, 14, 18, 28, 29, 33, 34, 37, 53, 56; pcna+, cdk1+) were identified according to putative marker genes (Figure 1—source data 5). On the other hand, among neuronal cells, the first segregation defined three classes of neurons (branch I, cerebellum and habenula; branch IIa, glutamatergic neurons, branch IIb, inhibitory neurons). Branch I consisted of granule cells (Clusters 19), torus longitudinals (Cluster 45), cranial ganglions (Clusters 19, 50, 63), dorsal and ventral habenula neurons (Clusters 35 and 59). Branch IIa included 22 subclasses of excitatory glutamatergic neurons (vgluta2a+-gad1b--glyt2-), whereas Branch IIb inhibitory neurons included 17 subclasses of GABAergic neurons (vgluta2a-/gad1b+) and 1 subclass of glycinergic neurons (vgluta2a--gad1b+-glyt2+, Cluster17, Figure 1—source data 5). Molecular classification of neuromodulator-type neurons To further examine the expression of neuromodulators at the whole-brain level, we performed the scRNA-seq analysis of neuromodulator neurons sorted from the whole brain of Tg (ETvmat2:GFP) transgenic fish (Figure 2A). Using this fishline, we could examine transcriptomes of DA neurons, 5-hydroxytryptamine (5-HT) neurons, and norepinephrinergic (NE) neurons (Wen et al., 2008). After the filtering procedures described above, we obtained a total of 5368 vmat2-expressing cells (Materials and methods). The analysis aggregated these cells into 22 clusters in t-SNE plots (Figure 2B), more than those found using whole-brain samples (two DA, one 5-HT, and three ChAT; Figure 1—source data 2). To further validate the stability of clusters, we used the Jaccard index, which showed that the majority of clusters (20/22) were stable using mean/median Jaccard index >0.6 as cutoff (Figure 2—figure supplement 1A). Figure 2 with 1 supplement see all Download asset Open asset Molecular classification of neuromodulator-type neurons. (A) The schematic showing the procedure of collecting single-cell transcriptomes of neuromodulator neurons with fluorescence-activated cell sorting (FACS). Using Tg (ETvmat2:GFP) fishline, we could isolate dopaminergic (DA), serotonergic (5-HT), and norepinephrinergic (NE) neurons. (B) The t-distributed stochastic neighbor embedding (t-SNE) plot of 5368 cells obtained from Tg (ETvmat2:GFP) fishline expressing monoaminergic neuromodulators, showing 22 clusters, each marked by a number, color-coding brain regions and cell type of each cluster. (C) Dot plot showing the expression of glutamatergic marker (vglut2a/vglut2b/vglu1/vglu3) and GABAergic marker (gad1b/gad2) in each of the 22 vmat2+ clusters in B. The neurotransmitter phenotypes were color-coded. Empty squares depict the ones with undefined neurotransmitter phenotypes. The average expression levels of these genes for all cells in each cluster were coded by the gray level. The percentage of cells expressing each gene within each cluster was coded by dot size. Figure 2—source data 1 The annotation of neuromodulator-type neuronal types with well-known markers. https://cdn.elifesciences.org/articles/68224/elife-68224-fig2-data1-v2.xlsx Download elife-68224-fig2-data1-v2.xlsx According to region-specific marker genes identified above and known marker genes for neuromodulator neurons, we assigned seven clusters (Fore: Clusters 18–19; sub-OT: Clusters 16, 20; OT: Cluster 22; and Clusters 17, 21 without specific regional identity) as DA neurons, 14 clusters (Hind: Cluster 10; sub-OT: Clusters 1, 2, 6, 7; and Clusters 3–5, 8–9, 11–14 without regional identity) as 5-HT neurons, one cluster (Hind: Cluster 15) as NE neurons (Figure 2—figure supplement 1C-D, Figure 2—source data 1). Further examination of these neuromodulator clusters for their expression of specific neurotransmitters showed that the majority of 5-HT (8/14) and DA clusters (5/7) expressed GABAergic markers gad1b/gad2 (Figure 2C). This result was further validated by Tg (ETvmat2:GFP::gad1b:gal4::uas:mCherry) (Figure 2—figure supplement 1E). Only three 5-HT clusters (Clusters 11–13) expressed glutamatergic marker vglut3, two DA clusters (Clusters 16, 17), and one NE cluster (Cluster 15) expressed glutamatergic markers, vglut2a and vglut2b (Figure 2C). These results are consistent with previous studies (Filippi et al., 2014). Besides, we also found three clusters (Clusters 24, 41, and 64) in whole brain showed choline and glutamate preferential co-expression (Figure 1—source data 2). Thus, our analysis provided a whole-brain characterization of the co-expression patterns of neurotransmitters and neuromodulators. In sum, DA and 5-HT neurons preferentially expressed GABAergic markers, whereas NE and ChAT neurons mostly expressed glutamatergic markers. The TF regulatory landscape of whole-brain glutamatergic/GABAergic neuron clusters In the hierarchical classification based on effector gene profiles (Figure 1F), glutamatergic/GABAergic clusters (n = 39) at the terminus pairs represented the ones with the most similar terminal features, termed ‘sister clusters’. In addition, the certainty of this hierarchical classification was verified by the bootstrap re-sampling analysis using pvclust v.2.0 (Figure 3—figure supplement 1A; Suzuki and Shimodaira, 2006). We identified 11 pairs of sister clusters, neurons in each pair exhibited the same neurotransmitter types (Figure 3—figure supplement 2A). To our surprise, neurons of each sister clusters could be from either the same (n = 6) or different (n = 5) brain regions (Figure 3—figure supplement 2A), which did not reflect the strong brain region preference. To further examine the TF profiles of these effector gene-based sister clusters, we classified glutamatergic (IIa) and inhibitory (IIb) neurotransmitter-type neurons using TF profiles. Notably, TF-based and effector gene-based trees were distinct in terms of matching node (only one matching node: Clusters 9/61) and tree distances (tree distance = 0.71, Figure 3—figure supplement 1B), suggesting that the effector gene-based sister clusters (Figure 3—figure supplement 2A) might express different TF profiles. We found out of 11 effector genes-based sister clusters, only one pairs could be found in TF-based sister clusters (Clusters 9/61, ‘matched pattern’, Figure 3—figure supplement 2B). And other 10 effector gene-based sister clusters were separated in TF-based classification, suggesting that neurons with similar terminal features mostly expressed different TF profiles (‘convergent pattern’; Figure 3—figure supplement 2C). Also, neurons in each of these 10 sister clusters could come from either the same (n = 5) or different (n = 5) brain regions, exhibiting no brain region preference (Figure 3—figure supplement 2C). Alternatively, we performed the population-level statistical analysis to compare the landscape of TF and effector gene expression accounting for the full spectrum of cell types rather than just the most similar sister clusters. For all glutamatergic/GABAergic neuronal clusters (n = 39), we calculated the distances between every two clusters (C392) based on either effector gene profiles or TF profiles, and then defined the pairs, which had the lowest 10% distances after ranking, as similar pair clusters (Figure 3A). Then, we defined paired clusters that were similar in both TF and effector gene profiles as ‘matched pattern’, those paired clusters that were similar in effector gene profiles but not in TF profiles as ‘convergent pattern’ (Figure 3B). The population-level analysis identified 19 pairs of effector gene-based similar pair clusters, 5 with matched pattern and 14 with convergent pattern (Figure 3—figure supplement 2D-G). Overall, similar pair clusters with either matched or convergent pattern identified by the population-level analysis showed an overlapping but distinct pattern with those identified by the hierarchical sister cluster analysis (Figure 3B). This discrepancy was likely due to the following facts: (1) In the population-level statistical analysis, we arbitrarily set the lowest threshold as a criterion to identify similar pair clusters. Thus, the levels of this threshold could influence the production of similar pair clusters. (2) In population-level statistical analysis, each cluster could use for multiple times, whereas in hierarchical sister cluster analysis, once a cluster was selected as a pair with another cluster, it could not be re-used again. To overcome this discrepancy, we intersected the results from hierarchical sister cluster analysis and population-level statistical analysis, and identified eight pairs of effector gene-based glutamatergic/GABAergic similar pair clusters, one with matched pattern and seven with convergent pattern (Figure 3B). Figure 3 with 4 supplements see all Download asset Open asset The transcription factor (TF) regulatory landscape in whole-brain neuronal clusters. (A) Schematic showing the strategies to assess the cluster similarity based on effector gene and TF profiles. We focused on clusters of whole-brain glutamatergic/GABAergic neurons and neuromodulator neurons. Similar pair clusters were identified by two strategies: the first strategy was based on hierarchical sister clusters. The second strategy was based on population-level statistical analysis, in which we calculated and ranked the distances of every two clusters from 39 neurotransmitter-type clusters (C392) and chose the ones with the lowest 10% distance as similar pair clusters. (B) Left: schematic showing the criteria of three patterns: pair clusters that were similar in both TF and effector gene profiles as ‘matched pattern’, those pair clusters that were similar in effector gene profiles but not in TF profiles as ‘convergent pattern’, and those pair clusters that were similar in TF profiles but not in effector gene profiles as ‘divergent pattern’. Right: the plot showing the number of each pattern using two strategies in A. The red dashed circle showing the number of cluster pairs with given pattern based on hierarchical sister cluster analysis; the black solid circle showing the number of cluster pairs with given pattern based on population-level statistical analysis. (C–D) Violin plots showing the expression of TFs (yellow) or effector genes (black) in glutamatergic and GABAergic similar pair clusters of convergent pattern. (E) The bar plot showing the proportions of different patterns for neuronal clusters with neurotransmitter or neuromodulator types. The numbers of each pattern were indicated. Fisher′s exact test was used to test the significant association of different patterns, p = 0.02564, *p < 0.05. (F–G) Violin plots showing the expression of TF profiles (yellow) or effector gene profiles (black) in neuronal clusters of divergent pattern. We further validated these patterns of paired neuronal clusters by subsampling of genes (80% of total either TFs or effector genes) and average statistics over 20 times to re-identify similar paired clusters based on either TFs or effector genes (Materials and methods). Notably, re-identified paired clusters of different patterns completely recapitulated those pairs identified using the population-level statistical analysis above (Figure 3—figure supplement 2L), indicating the robustness of these patterns. By combining sister cluster analysis and the population-level statistical analysis, we identified one paired clusters of ‘matched pattern’, seven paired clusters of ‘convergent pattern’ (Figure 3B). Here were some representative cases with the convergent pattern. The first case was a glutamatergic pair cluster from different brain regions: tectal glutamatergic Cluster 1 and hindbrain glutamatergic Cluster 31 shared effector gene profiles including camk2n1a/stmn4/cbln2b/olfm3a/cd63, but differentially expressed TF profiles, atf5b/bhlhe41/lhx1a and ddit3/cebpb/lef1, respectively (Figure 3C). The second case was GABAergic pair cluster from different brain regions: GABAergic clusters in the forebrain (Cluster 36) and the sub-OT (Cluster 16) shared effector gene profiles mcl1a/cbln1/tspan18a/dusp5/ncaldb, but differentially expressed TF profiles, tbr1b/foxg1a/barhl2 and otpa/otpb/six6a, respectively (Figure 3D). Consistently, each of the above TF profiles has previously been cha" @default.
W4205656760 created "2022-01-26" @default.
W4205656760 creator A5000079027 @default.
W4205656760 creator A5004036488 @default.
W4205656760 creator A5004307488 @default.
W4205656760 creator A5012813645 @default.
W4205656760 creator A5013444752 @default.
W4205656760 creator A5039215131 @default.
W4205656760 creator A5039787951 @default.
W4205656760 creator A5068400813 @default.
W4205656760 creator A5075819658 @default.
W4205656760 creator A5080703712 @default.
W4205656760 date "2021-07-16" @default.
W4205656760 modified "2023-09-29" @default.
W4205656760 title "Author response: The landscape of regulatory genes in brain-wide neuronal phenotypes of a vertebrate brain" @default.
W4205656760 doi "https://doi.org/10.7554/elife.68224.sa2" @default.
W4205656760 hasPublicationYear "2021" @default.
W4205656760 type Work @default.
W4205656760 citedByCount "0" @default.
W4205656760 crossrefType "peer-review" @default.
W4205656760 hasAuthorship W4205656760A5000079027 @default.
W4205656760 hasAuthorship W4205656760A5004036488 @default.
W4205656760 hasAuthorship W4205656760A5004307488 @default.
W4205656760 hasAuthorship W4205656760A5012813645 @default.
W4205656760 hasAuthorship W4205656760A5013444752 @default.
W4205656760 hasAuthorship W4205656760A5039215131 @default.
W4205656760 hasAuthorship W4205656760A5039787951 @default.
W4205656760 hasAuthorship W4205656760A5068400813 @default.
W4205656760 hasAuthorship W4205656760A5075819658 @default.
W4205656760 hasAuthorship W4205656760A5080703712 @default.
W4205656760 hasBestOaLocation W42056567601 @default.
W4205656760 hasConcept C104317684 @default.
W4205656760 hasConcept C127716648 @default.
W4205656760 hasConcept C169760540 @default.
W4205656760 hasConcept C2778961482 @default.
W4205656760 hasConcept C54355233 @default.
W4205656760 hasConcept C78458016 @default.
W4205656760 hasConcept C86803240 @default.
W4205656760 hasConceptScore W4205656760C104317684 @default.
W4205656760 hasConceptScore W4205656760C127716648 @default.
W4205656760 hasConceptScore W4205656760C169760540 @default.
W4205656760 hasConceptScore W4205656760C2778961482 @default.
W4205656760 hasConceptScore W4205656760C54355233 @default.
W4205656760 hasConceptScore W4205656760C78458016 @default.
W4205656760 hasConceptScore W4205656760C86803240 @default.
W4205656760 hasLocation W42056567601 @default.
W4205656760 hasOpenAccess W4205656760 @default.
W4205656760 hasPrimaryLocation W42056567601 @default.
W4205656760 hasRelatedWork W1972312173 @default.
W4205656760 hasRelatedWork W1972939812 @default.
W4205656760 hasRelatedWork W2057039167 @default.
W4205656760 hasRelatedWork W2089191032 @default.
W4205656760 hasRelatedWork W2099737709 @default.
W4205656760 hasRelatedWork W2110450140 @default.
W4205656760 hasRelatedWork W2469014158 @default.
W4205656760 hasRelatedWork W3034907954 @default.
W4205656760 hasRelatedWork W4238214718 @default.
W4205656760 hasRelatedWork W4301889343 @default.
W4205656760 isParatext "false" @default.
W4205656760 isRetracted "false" @default.
W4205656760 workType "peer-review" @default.