Phenotypic variation of transcriptomic cell types in mouse motor cortex

In the paper

“While we found that different t-types had different preferred morphologies (Extended Data Table 1), they showed substantial overlap, in agreement with the L5 Pvalb t-types themselves not having clear boundaries4 (Fig. 1d).”

What the sources say

• Partly supported

  Extended Data Fig. 3 reports pronounced co-clustering among certain Pvalb transcriptomic types, supporting the cited point that their boundaries can be unclear. It does not identify those types as L5 Pvalb types. The preferred morphologies and their overlap are presented as the citing authors’ own findings, not as findings attributed to this source. (LLM, conf=0.83)

  cites [4] Tasic, B. et al. Shared and distinct transcriptomic cell types across neocortical areas. Nature 563, 72–78 (2018). sourceExperimental — accuracy not yet measured

In the paper

“The resulting data set covered 77 out of the 90 neuronal t-types (Fig. 1a), with 73 t-types having at least one morphologically reconstructed neuron.”

In the paper

“The coverage was good for interneurons derived from the caudal and medial ganglionic eminences (CGE and MGE) and for excitatory neurons.”

In the paper

“We also showed that three previously described morphological types of L5 Pvalb cells5, as well as two morphological types of L5 Martinotti cells29,30, corresponded to different t-types.”

What the sources say

• Not checked yetcites [5] Jiang, X. et al. Principles of connectivity among morphologically defined cell types in adult neocortex. Science 350, aac9462 (2015). source
• Not checked yetcites [29] Muñoz, W., Tremblay, R., Levenstein, D. & Rudy, B. Layer-specific modulation of neocortical dendritic inhibition during active wakefulness. Science 355, 954–959 (2017). source
• Not checked yetcites [30] Nigro, M. J., Hashikawa-Yamasaki, Y. & Rudy, B. Diversity and connectivity of layer 5 somatostatin-expressing interneurons in the mouse barrel cortex. J. Neurosci. 38, 1622–1633 (2018). source

In the paper

“Similarly, Lamp5 Egln3_1 neurons had rather Vip- and Sst-like firing instead of the typical Lamp5 electrophysiology, and Vip Mybpc1 neurons often had Sst-like firing.”

In the paper

“The selected model used 25 genes with a 5-dimensional latent space and achieved a cross-validated R2 of 0.38.”

What the sources say

• Not checked yetcites [32] Kobak, D. et al. Sparse reduced-rank regression for exploratory visualization of paired multivariate data sets. R. Stat. Soc. 70, 980–1000 (2021). source

In the paper

“The observed phenotypes included most of the morphological and electrophysiological types of cortical neurons that have been described previously in mice and rats5,6,24, allowing us to link transcriptomic and morpho-electric descriptions (Extended Data Fig. 3, Supplementary File 1).”

What the sources say

• Not checked yetcites [5] Jiang, X. et al. Principles of connectivity among morphologically defined cell types in adult neocortex. Science 350, aac9462 (2015). source
• Not checked yetcites [6] Markram, H. et al. Reconstruction and simulation of neocortical microcircuitry. Cell 163, 456–492 (2015). source
• Not checked yetcites [24] Gouwens, N. W. et al. Classification of electrophysiological and morphological neuron types in the mouse visual cortex. Nat. Neurosci. 22, 1182–1195 (2019). source

In the paper

“For example, electrophysiological properties of the t-types within the Vip subclass varied continuously across the transcriptomic landscape; the membrane time constant, for instance, had its largest values close to the Sncg subclass and gradually decreased towards Vip Gpc3 (Fig. 4a).”

In the paper

“In summary, different transcriptomic families had largely distinct morpho-electric phenotypes, in agreement with them being well separated in the transcriptomic space4.”

What the sources say

• Not checked yetcites [4] Tasic, B. et al. Shared and distinct transcriptomic cell types across neocortical areas. Nature 563, 72–78 (2018). source

In the paper

“For example, L4/5 and L5 IT t-types that were transcriptomically close to the L2/3 IT t-types were located at the top of L5 close to the border between L2/3 and L5, whereas L5 IT t-types that were transcriptomically close to L6 IT t-types were located at the bottom of L5 close to the border with L6.”

In the paper

“The between-family separability was strongest when we had combined electrophysiological and morphological features into a joint representation (Fig. 3c, d), showing that these sets of properties are not redundant.”

In the paper

“These groups had distinct electrophysiological properties: for example, as expected, Pvalb neurons were characterized by high firing rates while Sst neurons had high values of the hyperpolarization sag and rebound (Fig. 2, right).”

In the paper

“However, we found that IT t-types did not form distinct groups for each cortical layer; instead, the soma depth and RNA expression varied continuously along a one-dimensional manifold (Fig. 4d), in agreement with parallel findings based on a spatial transcriptomics approach34.”

What the sources say

• Not checked yetcites [34] Zhang, M. et al. Spatially resolved cell atlas of the mouse primary motor cortex by MERFISH. Nature https://doi.org/10.1038/s41586-021-03705-x (2021). source

In the paper

“We were also able to identify a t-type, L4/5 IT_1, that was located on the boundary between L2/3 and L5 and probably corresponds to the quasi-L4 neurons described previously in motor cortex31.”

What the sources say

• Not checked yetcites [31] Yamawaki, N., Borges, K., Suter, B. A., Harris, K. D. & Shepherd, G. M. A genuine layer 4 in motor cortex with prototypical synaptic circuit connectivity. eLife 3, e05422 (2014). source

In the paper

“The correlation was also observed within multiple individual families and for many individual electrophysiological features (Extended Data Fig. 7).”

In the paper

“Our data sets are overall in good agreement (Extended Data Fig. 9) and together offer an unprecedented view of cell type variability in the neocortex.”

In the paper

“The shape of the axonal arbor showed continuous changes across the transcriptomic landscape (Fig. 4e): small shrub-like basket cells, horizontally elongated basket cells, and vertically elongated classical basket cells were located in different corners of the t-SNE embedding, with intermediate morphologies in between.”

In the paper

“Most t-types could be unambiguously placed into the correct family (Fig. 5a), but some t-types were in between two families.”

In the paper

“By contrast, Vip Mybpc1_2 was marked as having high electrophysiological variability and indeed had high variance in input resistance, membrane time constant, and rebound (Extended Data Fig. 4).”

In the paper

“Similarly, Sst Pvalb Calb2 cells had high variability in terms of the maximum firing rate, but high-firing cells were mostly grouped in one part of the transcriptomic landscape (Fig. 5d).”

In the paper

“We observed the same in the Sst subclass, which is known to be transcriptomically4 and morpho-electrically29,30,33 diverse in L5. Here we also found that morpho-electric properties varied continuously across the transcriptomic landscape, with neighbouring t-types consistently showing similar morphologies and similar rebound values (Fig. 4b).”

What the sources say

• Not checked yetcites [4] Tasic, B. et al. Shared and distinct transcriptomic cell types across neocortical areas. Nature 563, 72–78 (2018). source
• Not checked yetcites [29] Muñoz, W., Tremblay, R., Levenstein, D. & Rudy, B. Layer-specific modulation of neocortical dendritic inhibition during active wakefulness. Science 355, 954–959 (2017). source
• Not checked yetcites [30] Nigro, M. J., Hashikawa-Yamasaki, Y. & Rudy, B. Diversity and connectivity of layer 5 somatostatin-expressing interneurons in the mouse barrel cortex. J. Neurosci. 38, 1622–1633 (2018). source
• Not checked yetcites [33] Naka, A. et al. Complementary networks of cortical somatostatin interneurons enforce layer specific control. eLife 8, e43696 (2019). source

In the paper

“Within each class, cells were strongly segregated by the soma depth, with excitatory cells forming mostly a one-dimensional manifold.”

In the paper

“The resulting cDNA yield was not correlated with the hold time (Spearman correlation −0.01).”

In the paper

“We were unable to find any labelled cells in MOp in the Gnb4-Cre mice: all labelled cells were far outside of MOp and close to the claustrum46.”

What the sources say

• Not checked yetcites [46] Peng, H. et al. Morphological diversity of single neurons in molecularly defined cell types. Nature https://doi.org/10.1038/s41586-021-03941-1 (2021). source

In the paper

“The addition of the inhibitor resulted in an increase in osmolarity to the desired value of 315–320 mOSM without a further dilution17.”

What the sources say

• Not checked yetcites [17] Cadwell, C. R. et al. Multimodal profiling of single-cell morphology, electrophysiology, and gene expression using Patch-seq. Nat. Protocols 12, 2531 (2017). source

In the paper

“A substantial portion of the RNA extracted from the neurons was nascent and contained intronic sequences.”

In the paper

“We detected 42,466 genes, including pseudogenes and annotated non-coding segments, in at least one cell.”

In the paper

“One hundred and ten cells mapped to the Sst subclass, 43 to IT, 12 to ET, 10 to Pvalb, and 2 to NP.”

In the paper

“175 cells were assigned to L5 in the post hoc analysis (see below).”

In the paper

“Instead, neurons form a hierarchy that consists of distinct non-overlapping branches at the level of families, but can form continuous and correlated transcriptomic and morpho-electrical landscapes within families.”

• Also stated in: § Discussion · plain

  “In this metaphor, neurons follow a hierarchy consisting of distinct, non-overlapping branches at the level of families (large leaves), but with a spectrum of cells forming continuous and correlated transcriptomic and morpho-electrical landscapes within each leaf.”

In the paper

“Cells that showed poor mapping (owing to a low read count or excessive RNA contamination) were excluded (Extended Data Fig. 2), leaving 1,227 neurons for further analysis (817 inhibitory, 410 excitatory; 369 and 269 with morphological reconstructions, respectively).”

In the paper

“However, to understand the roles of t-types in cortical computations, it is necessary to know their anatomy, connectivity, and electrophysiology7.”

What the sources say

• Not checked yetcites [7] Zeng, H. & Sanes, J. R. Neuronal cell-type classification: challenges, opportunities and the path forward. Nat. Rev. Neurosci. 18, 530–546 (2017). source

In the paper

“In total, we performed whole-cell recordings from more than 2,000 cells, of which 1,329 cells (from 266 mice) passed initial quality control.”

In the paper

“The mRNA of these 1,329 cells was sequenced, yielding on average 1.3 million exonic and 0.7 million intronic reads (medians; mean ± s.d. on a log10 scale: 6.0 ± 0.6 and 5.6 ± 0.8, respectively) and 9,100 ± 3,500 (mean ± s.d.) detected genes per cell (Extended Data Fig. 2).”

In the paper

“High-throughput single-cell sequencing techniques have identified dozens of types of neuron on the basis of their transcriptional profiles3,4,14,15, but linking transcriptomically defined cell types (t-types) to their phenotypes has remained a major challenge16.”

What the sources say

• Not checked yetcites [16] Huang, Z. J. & Paul, A. The diversity of GABAergic neurons and neural communication elements. Nat. Rev. Neurosci. 20, 563–572 (2019). source

In the paper

“The current view is that a neuronal cell type is characterized by a common genetic profile that gives rise to distinct physiological and anatomical properties, including patterns of connectivity7,9.”

What the sources say

• Not checked yetcites [7] Zeng, H. & Sanes, J. R. Neuronal cell-type classification: challenges, opportunities and the path forward. Nat. Rev. Neurosci. 18, 530–546 (2017). source
• Not checked yetcites [9] Masland, R. H. Neuronal cell types. Curr. Biol. 14, R497–R500 (2004). source

In the paper

“Of these neurons, 646 had sufficient staining for their morphologies to be reconstructed.”

In the paper

“Cortical neurons exhibit extreme diversity in gene expression as well as in morphological and electrophysiological properties1,2.”

What the sources say

• Not checked yetcites [1] Harris, K. D. & Shepherd, G. M. The neocortical circuit: themes and variations. Nat. Neurosci. 18, 170–181 (2015). source
• Not checked yetcites [2] Tremblay, R., Lee, S. & Rudy, B. GABAergic interneurons in the neocortex: from cellular properties to circuits. Neuron 91, 260–292 (2016). source

In the paper

“High-throughput single-cell sequencing techniques have identified dozens of types of neuron on the basis of their transcriptional profiles3,4,14,15, but linking transcriptomically defined cell types (t-types) to their phenotypes has remained a major challenge16.”

What the sources say

• Not checked yetcites [3] Tasic, B. et al. Adult mouse cortical cell taxonomy revealed by single cell transcriptomics. Nat. Neurosci. 19, 335–346 (2016). source
• Not checked yetcites [4] Tasic, B. et al. Shared and distinct transcriptomic cell types across neocortical areas. Nature 563, 72–78 (2018). source
• Not checked yetcites [14] Zeisel, A. et al. Cell types in the mouse cortex and hippocampus revealed by single-cell RNA-seq. Science 347, 1138–1142 (2015). source
• Not checked yetcites [15] Zeisel, A. et al. Molecular architecture of the mouse nervous system. Cell 174, 999–1014.e22 (2018). source

In the paper

“Our results suggest that although all deep NGCs belong to the Lamp5 subclass, some are derived from the CGE and some from the MGE, as in the hippocampus25–27.”

What the sources say

• Not checked yetcites [25] Tricoire, L. et al. Common origins of hippocampal Ivy and nitric oxide synthase expressing neurogliaform cells. J. Neurosci. 30, 2165–2176 (2010). source
• Not checked yetcites [26] Pelkey, K. A. et al. Hippocampal GABAergic inhibitory interneurons. Physiol. Rev. 97, 1619–1747 (2017). source
• Not checked yetcites [27] Harris, K. D. et al. Classes and continua of hippocampal CA1 inhibitory neurons revealed by single-cell transcriptomics. PLoS Biol. 16, e2006387 (2018). source

In the paper

“Consistent with our interpretation, recent transcriptomic and anatomical studies have argued that neurons in hippocampus, striatum, and cerebellum can be better described as forming partially continuous manifolds27,37–39.”

What the sources say

• Not checked yetcites [27] Harris, K. D. et al. Classes and continua of hippocampal CA1 inhibitory neurons revealed by single-cell transcriptomics. PLoS Biol. 16, e2006387 (2018). source
• Not checked yetcites [37] Muñoz-Manchado, A. B. et al. Diversity of interneurons in the dorsal striatum revealed by single-cell RNA sequencing and PatchSeq. Cell Rep. 24, 2179–2190.e7 (2018). source
• Not checked yetcites [38] Stanley, G. et al. Continuous and discrete neuron types of the adult murine striatum. Neuron 105, 688–699.e8 (2019). source
• Not checked yetcites [39] Wang, W. X. & Lefebvre, J. L. Morphological pseudotime ordering and fate mapping reveals diversification of cerebellar inhibitory interneurons. Preprint at https://doi.org/10.1101/2020.02.29.971366 (2020). source

In the paper

“Although we cannot fully exclude the possibility that some of this variability can be attributed to technical challenges of Patch-seq or to factors such as the exact spatial location of the cell within motor cortex36, there are clear cases in our data that suggest that within-type morpho-electric variability is related to within-type transcriptomic variability.”

What the sources say

• Not checked yetcites [36] Cembrowski, M. S. et al. Spatial gene-expression gradients underlie prominent heterogeneity of ca1 pyramidal neurons. Neuron 89, 351–368 (2016). source

In the paper

“Thus, while overall transcriptomic family was highly predictive of the cell phenotype, some t-types exhibited properties similar to those of another transcriptomic family.”

In the paper

“Previous studies using transgenic lines had shown that morpho-electric properties within these families can be highly variable5,24.”

What the sources say

• Not checked yetcites [5] Jiang, X. et al. Principles of connectivity among morphologically defined cell types in adult neocortex. Science 350, aac9462 (2015). source
• Not checked yetcites [24] Gouwens, N. W. et al. Classification of electrophysiological and morphological neuron types in the mouse visual cortex. Nat. Neurosci. 22, 1182–1195 (2019). source

In the paper

“A parallel Patch-seq study of the inhibitory neurons in the mouse visual cortex45 focused on isolating multimodal neural types (‘met-types’) but also often observed continuous variation.”

What the sources say

• Not checked yetcites [45] Gouwens, N. W. et al. Integrated morphoelectric and transcriptomic classification of cortical GABAergic cells. Cell 183, 1–19 (2020). source

In the paper

“No statistical methods were used to predetermine sample size.”

In the paper

“Within-type distributions of soma depths (Fig. 1b) agreed well with previous data4 and with the layer-specific nomenclature of excitatory t-types, confirming the validity of our t-type assignment.”

What the sources say

• Not checked yetcites [4] Tasic, B. et al. Shared and distinct transcriptomic cell types across neocortical areas. Nature 563, 72–78 (2018). source

In the paper

“Similarly, cortical studies have identified many intermediate cells with uncertain t-type assignments3,4.”

What the sources say

• Not checked yetcites [3] Tasic, B. et al. Adult mouse cortical cell taxonomy revealed by single cell transcriptomics. Nat. Neurosci. 19, 335–346 (2016). source
• Not checked yetcites [4] Tasic, B. et al. Shared and distinct transcriptomic cell types across neocortical areas. Nature 563, 72–78 (2018). source

In the paper

“Second, as the RNA extraction process may have interfered with biocytin diffusion17 and as MOp is quite thick, it was difficult to recover complete morphologies of some groups of neurons, such as deep L5 Martinotti cells with thin long axons that reach all the way to L1.”

What the sources say

• Not checked yetcites [17] Cadwell, C. R. et al. Multimodal profiling of single-cell morphology, electrophysiology, and gene expression using Patch-seq. Nat. Protocols 12, 2531 (2017). source

In the paper

“The experiments were not randomized and investigators were not blinded to allocation during experiments and outcome assessment, unless otherwise stated.”

In the paper

“This is in agreement with the idea that on a fine within-family scale, both transcriptomic and morpho-electric landscapes are continuous rather than discrete.”

In the paper

“We obtained high-quality electrophysiological recordings and extracted electrophysiological features of 184 cells.”

In the paper

“The investigators were blinded to the cell type of each sample during library construction and sequencing.”

In the paper

“Two inhibitory neurons were left out because some of the morphometric statistics could not be extracted owing to insufficient dendritic recovery; this left 367 inhibitory neurons (with 50 morphometric features) and 269 excitatory neurons (with 35 morphometric features).”

In the paper

“We obtained soma depth information for 1,284 neurons out of 1,329 (45 neurons were mistakenly recorded using a solution with insufficient biocytin concentration, and we could measure soma depths for only 2 of those; for 2 other neurons the measurements could not be carried out because the slices were lost).”

In the paper

“Samples containing less than around 1 ng total cDNA (in the 15 μl final volume) or with an average size less than 1,500 bp were typically not sequenced (with some occasional exceptions).”

In the paper

“We did not use any axonal features for the excitatory cells because only a small part of the axon could typically be reconstructed.”

In the paper

“We sequenced in total 185 neurons, obtained from 8 adult mice (7 Sst-Cre/Ai9 and 1 Pvalb-Cre/Ai9), of which 177 neurons passed the transcriptomic quality control and got a t-type assignment (see below).”

In the paper

“We took special care to exclude reconstructions of all neurons that showed any signs of damage, lack of contrast, or poor overall staining.”

In the paper

“In addition, one cell was probably located outside MOp, based on the slice anatomy, and was excluded as well.”

In the paper

“For all statistical analysis we used 17 features out of the extracted 29, excluding features that were equal to zero for many cells (afterdepolarization, burstiness, rebound number of APs, sag area, sag time, wildness), two Fano factor features that were highly correlated with the corresponding coefficient of variation features (AP Fano factor, ISI Fano factor) and another measure of latency that”

In the paper

“This is effectively a leave-one-out cross-validation procedure.”

In the paper

“To prevent overlap between these two populations, we added a small constant value of 0.25 to the excitatory block-diagonal block, leading to the strong excitatory–inhibitory separation in Fig. 3b.”

In the paper

“We used the entropies of a Leiden clustering35 as an alternative way to approach the same question.”

What the sources say

• Not checked yetcites [35] Traag, V. A., Waltman, L. & van Eck, N. J. From Louvain to Leiden: guaranteeing well-connected communities. Sci. Rep. 9, 5233 (2019). source

In the paper

“We did not use any features measuring morphological properties in the slice depth (y) direction because of possible slice shrinkage artefacts.”

In the paper

“Forty-five sequenced cells were mistakenly recorded using a solution with a much smaller concentration of biocytin, and their morphologies could not be recovered.”

In the paper

“We used bootstrapping over genes to assess the confidence of each t-type assignment.”

In the paper

“To provide a sensible baseline for the range of possible normalized total variances in a population of morpho-electrically homogeneous types, we used a clustering analysis.”

In the paper

“The remaining 1,232 cells passed quality control and entered the second round.”

In the paper

“The distributions of the remaining features were visually checked for outliers and for meaningful variation between transcriptomic types, leading to a further exclusion of three features for the inhibitory cells.”

In the paper

“Special care was taken to ensure that the seal between the pipette and the cell membrane remained intact to reduce contamination from the extracellular environment.”

In the paper

“We did not use the Sncg and NP families owing to insufficient coverage in our data set (Fig. 1).”

In the paper

“We used the following quality control criteria: (1) seal resistance value >1 GΩ before achieving whole-cell configuration; (2) access resistance <30 MΩ.”

In the paper

“In each case, we excluded features with coefficient of variation below 0.25 (among the features with only positive values).”

In the paper

“The final number of cells with t-type assignment was 1,227.”

In the paper

“We reduced each set to five PCs, discarded PC1 (it was strongly correlated with the normalized soma depth and made the resulting embedding strongly influenced by the soma depth), and scaled the PCs by the standard deviation of the respective PC2.”

In the paper

“the three large transcriptomic groups (CGE-derived interneurons, MGE-derived interneurons, and excitatory neurons) that are perfectly separated from each other with no transcriptomically intermediate cells4.”

What the sources say

• Not checked yetcites [4] Tasic, B. et al. Shared and distinct transcriptomic cell types across neocortical areas. Nature 563, 72–78 (2018). source

In the paper

“This shrinkage was not compensated for in our analysis.”

In the paper

“Consistently with previous studies, tissue shrinkage due to the fixation and staining procedures was about 10–20%5,28,49.”

What the sources say

• Not checked yetcites [5] Jiang, X. et al. Principles of connectivity among morphologically defined cell types in adult neocortex. Science 350, aac9462 (2015). source
• Not checked yetcites [28] Scala, F. et al. Layer 4 of mouse neocortex differs in cell types and circuit organization between sensory areas. Nat. Commun. 10, 4174 (2019). source
• Not checked yetcites [49] Markram, H., Lübke, J., Frotscher, M., Roth, A. & Sakmann, B. Physiology and anatomy of synaptic connections between thick tufted pyramidal neurones in the developing rat neocortex. J. Physiol. (Lond.) 500, 409–440 (1997). source

In the paper

“We made sure that the measured electrophysiological properties of these cells were not systematically different from those of the the other sequenced cells.”

In the paper

“Four features were log-transformed to make their distribution more Gaussian-like: AP coefficient of variation, ISI coefficient of variation, ISI adaptation index, and latency.”

In the paper

“Cells meeting any of the following exclusion criteria were declared low quality and did not get a t-type assignment (Extended Data Fig. 2e): cells with the highest correlation below 0.4 (78 cells); cells that would be assigned to non-neural t-types, presumably owing to RNA contamination51 (14 cells; see also Extended Data Fig. 2j–n); cells with the highest correlation less than 0.02 above the max”

In the paper

“These four cells were excluded from all analyses and visualizations (as if they did not pass the transcriptomic quality control).”

In the paper

“The confidences obtained with seven reference data sets agreed well with each other (Extended Data Fig. 2i) and were averaged to obtain the consensus confidence.”

In the paper

“Subsampling was done to eliminate a possible bias due to the t-type abundance.”