Change text definitionintratelencepalic-projecting glutamatergic neurons in CL. Consider including their morphological and transcriptomic characteristics and just insert the layer of origin in the "comment" section

[AvolaAmg]

Change text definition for intratelencepalic-projecting glutamatergic neurons in CL. Mention transcriptomic distinction, include morphological and electrophysiological definition where possible, mention soma location where possible. Add standard transcriptomic reference data from CellxGene census.

[AvolaAmg]

I will review the text definition for the CL terms listed below and see if I can include more morphological, electrophysiological, transcriptomic informations. In the comment section, I will mention that historically these cells were defined based on their layer positioning, but this definition has changed due to novel transcriptomic informations.

This is the list of CL terms I will review for this PR:

L2 intratelencephalic projecting glutametergic neuron L2/3 intratelencephalic projecting glutamatergic neuron L2/3 intratelencephalic projecting glutamatergic neuron of the primary motor cortex L3 intratelencephalic projecting glutamatergic neuron L4 intratelencephalic projecting glutamatergic neuron L4/5 intratelencephalic projecting glutamatergic neuron L4/5 intratelencephalic projecting glutamatergic neuron of the primary motor cortex L5 intratelencephalic projecting glutamatergic neuron L5 intratelencephalic projecting glutamatergic neuron of the primary motor cortex L6 intratelencephalic projecting glutamatergic neuron L6 intratelencephalic projecting glutamatergic neuron of the primary motor cortex L6 intratelencephalic projecting Car3 glutamatergic neuron L6 intratelencephalic projecting glutamatergic neuron of the primary motor cortex L5/6 near-projecting glutamatergic neuron L5/6 near-projecting glutamatergic neuron of the primary motor cortex

[AvolaAmg]

TEXT DEFINITIONS TO CHANGE

L2 intratelencephalic projecting glutametergic neuron

Old definition: An intratelencephalic-projecting glutamatergic neuron with a soma found in cortical layer 2.

new definition: A transcriptomically distinct intratelencephalic-projecting glutamatergic neuron with a soma found in cortical layer 2 in the Mmus. The standard transcriptomic reference for this cell type can be found in the CellxGene census under the collection: "A taxonomy of transcriptomic cell types across the isocortex and hippocampal formation", dataset: "Single-cell RNA-seq for all cortical & hippocampal regions (10x)", Author Categories: "subclass_label", L2 IT ENT EN, L2 IT ENTm EN.

L2/3 intratelencephalic projecting glutamatergic neuron

Old definition: An intratelencephalic-projecting glutamatergic neuron with a soma found in cortical layer 2/3.

New definition: A transcriptomically distinct intratelencephalic-projecting glutamatergic neuron with a soma found between cortical layer 2-4. This intratelencephalic-projecting glutamatergic neuron has thin-tufted apical dendrites and extends its axonal projection into L5 in the neocortex. This neuronal type has a hyperpolarised resting membrane potential. The standard transcriptomic reference data for this cell type can be found on the CellxGene census under the collection: "Transcriptomic cytoarchitecture reveals principles of human neocortex organization", dataset: "Supercluster: IT-projecting excitatory neurons", Author Categories: "CrossArea_cluster", clusters L2/3 IT_1-6. https://www.nature.com/articles/nn.3917#Tab2

Comment:

Hystorically, the L2/3 intratelencephalic projectic glutametergic neuron was known to have its soma in cortical layer 2/3. MERFISH data showed that this intratelencephalic-projecting glutametergic neuron can have its soma in layer 2/3, 4B, 4C. The position of the soma in layer 4b and 4C is less frequent for this neuronal type in comparison to cortical layer 2/3.

L2/3 intratelencephalic projecting glutamatergic neuron of the primary motor cortex

Old Definition An intratelencephalic-projecting glutamatergic neuron with a soma found in cortical layer 2/3 of the primary motor cortex.

New defintion:

A transcriptomically distinct intratelencephalic-projecting glutamatergic neuron with a soma found in cortical layer 2/3 of the primary motor cortex. The standard transcriptomic reference for this cell type can be found in the CellxGene census under the collection "An integrated transcriptomic and epigenomic atlas of mouse primary motor cortex cell types", dataset: "An integrated transcriptomic and epigenomic atlas of mouse primary motor cortex cell types", Author_categories: "BICCN_subclass_label", L2/3 IT.

L3 intratelencephalic projecting glutamatergic neuron

Old definition: An intratelencephalic-projecting glutamatergic neuron with a soma found in cortical layer 3.

New definition: A transcriptomically distinct intratelencephalic-projecting glutamatergic neuron with a soma found in cortical layer 3. The standard transcriptomic reference for this cell type can be found in the CellxGene census under the collection: "A taxonomy of transcriptomic cell types across the isocortex and hippocampal formation", dataset: "Single-cell RNA-seq for all cortical & hippocampal regions (10x)", Author Categories: "subclass_label", L3 IT ENT.

L4 intratelencephalic projecting glutamatergic neuron

Old definition: An intratelencephalic-projecting glutamatergic neuron with a soma found in cortical layer 4.

New Definition: A transcriptomically distinct intratelencephalic-projecting glutamatergic neuron with a soma found in cortical layer 3-4. This neuron type can have a pyramidal, star-pyramidal or spiny stellate morphology and projects its output to L2/3 and L5A/B. The standard transcriptomic reference data for this cell type can be found on the CellxGene census under the collection: "Transcriptomic cytoarchitecture reveals principles of human neocortex organization", dataset: "Supercluster: IT-projecting excitatory neurons", Author Categories: "CrossArea_cluster", L4 IT_1-6.

Comment

Historically, the L4 intratelencephalic projectic glutametergic neuron was known to have its soma in cortical layer 4. MERFISH data showed that this intratelencephalic-projecting glutametergic neuron can have its also soma in layer 3.

L4/5 intratelencephalic projecting glutamatergic neuron

Old Definition: An intratelencephalic-projecting glutamatergic with a soma located in cortical layer 4/5.

New Definition:

A transcriptomically distinct intratelencephalic-projecting glutamatergic with a soma located in cortical layer 4/5 in the Mmus. The standard transcriptomic reference data for this cell type can be found in CellxGene census under the collection: "A taxonomy of transcriptomic cell types across the isocortex and hippocampal formation", dataset: "Single-cell RNA-seq for all cortical & hippocampal regions (10x)", Author Categories: "subclass_label", L4/5 IT CTX.

L4/5 intratelencephalic projecting glutamatergic neuron of the primary motor cortex

Old Definition: An intratelencephalic-projecting glutamatergic with a soma located in upper L5 of the primary motor cortex. These cells have thin untufted apical dendrites.

PERSONAL NOTES: Add this paper https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8113357/ and this paper https://pubmed.ncbi.nlm.nih.gov/33184512/ and potentially both references transcriptomically one of these papers have also genes as markers, I might add these in the comments, do not feel confident to put those in the definition.

New Definition: A transcriptomically distinct intratelencephalic-projecting glutamatergic with a soma located in upper L5 of the primary motor cortex. These cells have untufted or tufted apical dendrites. The standard transcriptomic reference data can be found in the CellxGene census under the collection: "Phenotypic variation of transcriptomic cell types in mouse motor cortex", dataset: "Excitatory neurons integrated with 10X sequencing MOp data", Author Categories: "BICCN_subclass_label", L4/5 IT.

Comment:

In Mmus, a L4/5 intratelencephalic projecting glutamatergic neuron of the primary motor cortex expresses Cux2, Rspo1, Htrc2, Rorb. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8494634/

L5 intratelencephalic projecting glutamatergic neuron

New Definition: An intratelencephalic-projecting glutamatergic neuron with a soma found in cortical layer 5. The standard transcriptomic reference data for this cell type can be found on the CellxGene census under the collection: "Transcriptomic cytoarchitecture reveals principles of human neocortex organization", dataset: "Supercluster: IT-projecting excitatory neurons", Author Categories: "CrossArea_cluster", L5 IT_1-6.

L5 intratelencephalic projecting glutamatergic neuron of the primary motor cortex

New Definition A transcriptomically distinct intratelencephalic-projecting glutamatergic neuron with a soma found in cortical layer 5 of the motor cortex. This neuron type has thin-tufted dendrites and presents a regular firing rate. The standard transcriptomic reference data for this cell type can be found on the CellxGene census under the collection: "Phenotypic variation of transcriptomic cell types in mouse motor cortex", dataset: "Excitatory neurons integrated with 10X sequencing MOp data", Author Categories: "BICCN_subclass_label", L5 IT. https://pubmed.ncbi.nlm.nih.gov/33184512/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8494634/ https://www.nature.com/articles/s41586-021-03705-x

L6 intratelencephalic projecting glutamatergic neuron

New Definition An transcriptomically distinct intratelencephalic-projecting glutamatergic neuron with a soma found in cortical layer 6. The standard transcriptomic reference data for this cell type can be found on the CellxGene census under the collection: "Transcriptomic cytoarchitecture reveals principles of human neocortex organization", dataset: "Supercluster: IT-projecting excitatory neurons", Author Categories: "CrossArea_cluster", L6 IT_1-2.

[L6 intratelencephalic projecting glutamatergic neuron of the primary motor cortex]()

New Definition An transcriptomically distinct intratelencephalic-projecting glutamatergic neuron with a soma found in L6 of the primary motor cortex. These cells are short untufted pyramidal cells, which could be stellate or inverted. The standard transcriptomic reference data for this cell type can be found on the CellxGene census under the collection: "Phenotypic variation of transcriptomic cell types in mouse motor cortex", dataset: "Excitatory neurons integrated with 10X sequencing MOp data", Author Categories: "BICCN_subclass_label", L6 IT.

https://www.nature.com/articles/s41586-021-03705-x https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8494634/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8113357/ New synonym https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8113357/ L6 IT t-type

L6 intratelencephalic projecting Car3 glutamatergic neuron

New Definition A transcriptomically distinct intratelencepalic-projecting glutamatergic neuron that expresses Car3 with a soma found in L6 . The standard transcriptomic reference data for this cell type can be found on the CellxGene census under the collection: "Transcriptomic cytoarchitecture reveals principles of human neocortex organization", dataset: "Supercluster: IT-projecting excitatory neurons", Author Categories: "CrossArea_cluster", L6 IT Car3.

https://link.springer.com/article/10.1007/s00424-024-02923-2

L5/6 near-projecting glutamatergic neuron

New Definition A transcriptomically distinct near-projecting glutamatergic neuron with a soma found in cortical layer 5/6. The standard transcriptomic reference data for this cell type can be found on the CellxGene census under the collection: "Transcriptomic cytoarchitecture reveals principles of human neocortex organization", dataset: "Supercluster: Deep layer (non-IT) excitatory neurons", Author Categories: "CrossArea_cluster", L5/6 NP_1-6.

[L5/6 near-projecting glutamatergic neuron of the primary motor cortex]()

New Definition A transcriptomically distinct near-projecting glutamatergic neuron with a soma found in layer 5/6 of the primary motor cortex. The standard transcriptomic reference data for this cell type can be found on the CellxGene census under the collection: "Phenotypic variation of transcriptomic cell types in mouse motor cortex", dataset: "Excitatory neurons integrated with 10X sequencing MOp data", Author Categories: "BICCN_subclass_label", L5/6 NP neuron.

Get rid of the first reference because it is in visual cortex PMID: 31209381

Add this reference https://www.nature.com/articles/s41586-020-2907-3

[AvolaAmg]

@aleixpuigb @dosumis Please review when you can :) and I will turn this ticket into a PR.

[dosumis]

I think this is a good start.

Some thoughts:

• For all transcriptomic references we should generate taxonomies from datasets & use the cell set IDs from these to generate formal links to the reference cell sets in the ontology.
• For CL this should be restricted to a small set of agreed reference datasets, but we can be broader in PCL (as we already are)
• We need to discuss whether we can/should reference multiple cells sets in a definition (we don't have formal linking mechanisms to cover this right now.

[dosumis]

For CL, I think we need to concentrate on cell types defined in Jorstad for now (crossArea_subclass)

e.g. => something like

A transcriptomically distinct near-projecting glutamatergic neuron with a soma found in cortical layer 5/6. The standard transcriptomic reference data for this cell type can be found on the CellxGene census under the collection: "Transcriptomic cytoarchitecture reveals principles of human neocortex organization", dataset: "Supercluster: Deep layer (non-IT) excitatory neurons", Author Categories: "CrossArea_subclass", L5/6 NP.

(But we may also be able to add NS-Forest markers.

[AvolaAmg]

Dear David,

Thanks for your comments!

For all transcriptomic references we should generate taxonomies from datasets & use the cell set IDs from these to generate formal links to the reference cell sets in the ontology.

Waiting to see what Jason and Brian say

For CL this should be restricted to a small set of agreed reference datasets, but we can be broader in PCL (as we already are) We need to discuss whether we can/should reference multiple cells sets in a definition (we don't have formal linking mechanisms to cover this right now.

Agreed, I think we need to run some bioinformatics analysis and justify why we believe those cell types can be referenced by two datasets. We could consider to use one dataset as the main reference in the text definition and have additional dataset where the cell types also appears in the comment section. For the Analysis I suggest running an integration across the two datasets and a person correlation to see how transcriptomically related the same cell type is across the two datasets.

For CL, I think we need to concentrate on cell types defined in Jorstad for now (crossArea_subclass)

For now, I will add on CL the new text definitions for Jorstad crossArea_subclass cell types. I will turn this into a PR.

[dosumis]

For all transcriptomic references we should generate taxonomies from datasets & use the cell set IDs from these to generate formal links to the reference cell sets in the ontology.

Waiting to see what Jason and Brian say.

I think we should do this whatever the response. The taxonomy stores cell IDs already, so this gives us a formal mechanism for reference with sufficient information to resolve the cell set - whatever technical mechanisms we put in place for resolution.

We can work with @hkir-dev on design patterns to add the relevant links to cell sets defined in taxonomies.

[scheuerm]

We need to discuss whether we can/should reference multiple cells sets in a definition (we don't have formal linking mechanisms to cover this right now.

Agreed, I think we need to run some bioinformatics analysis and justify why we believe those cell types can be referenced by two datasets. We could consider to use one dataset as the main reference in the text definition and have additional dataset where the cell types also appears in the comment section. For the Analysis I suggest running an integration across the two datasets and a person correlation to see how transcriptomically related the same cell type is across the two datasets.

We have some experience using a computational approach call FR-Match, to link cell types together from multiple datasets. There are a couple of other tools developed by other groups for doing this as well. Here's a link to our paper, where we compare how they perform - https://www.nature.com/articles/s41598-022-14192-z

[dosumis]

For the Analysis I suggest running an integration across the two datasets and a person [Pearson (?)] correlation to see how transcriptomically related the same cell type is across the two datasets.

We have some experience using a computational approach call FR-Match, to link cell types together from multiple datasets.

Let's see if we have resources. This could be a good place we could collaborate. Scores could be very useful to add to CL-KG.